A correlative IR, MS, 1H, 13C and 15 NMR and theoretical study of 4-arylthiazol-2(3H)-ones

Article abstract of DOI:10.1039/b106322g

Sixteen 4-arylthiazol-2(3H)-ones (3) were synthesised by cyclisation of α-thiocyanatoacetophenones (1) in acid solution. They appear to prefer greatly the oxo tautomeric forms. In CCI₄ solution an equilibrium between the free C=O bond and a "di

Full text of DOI:10.1039/b106322g

View Article Online / Journal Homepage / Table of Contents for this issue
1
13
15
A correlative IR, MS, H, C and N NMR and theoretical study
of 4-arylthiazol-2(3H)-ones†
Kalevi Pihlaja,*a Vladimir Ovcharenko,a Erkki Kolehmainen,b Katri Laihia,b
Walter M. F. Fabian,c Heinz Dehne,d Alexander Perjéssy,f Marion Kleist,d Joachim Tellere
and Zora Susteková
2
f
ˇ
a Department of Chemistry, University of Turku, FIN-20014 Turku, Finland
b Department of Chemistry, University of Jyväskylä, FIN-40351 Jyväskylä, Finland
c Institute of Chemistry, Karl Franzens University, Heinrichstr. 28, A-8010 Graz, Austria
d Department of Chemistry, University of Rostock, Buchbinderstr. 9, D-18051 Rostock,
Germany
e Micromod Partikeltechnologie GmbH, Friedrich-Barnewitz-Str. 4,
D-18119 Rostock-Warnemünde, Germany
f Department of Organic Chemistry and Institute of Chemistry, Faculty of Natural Sciences,
Comenius University, Mlynská dolina CH-2, SK-842 15 Bratislava, Slovak Republic
Received (in Cambridge, UK) 17th July 2001, Accepted 21st November 2001
First published as an Advance Article on the web 11th January 2002
Sixteen 4-arylthiazol-2(3H )-ones (3) were synthesised by cyclisation of α-thiocyanatoacetophenones (1) in acid
solution. They appear to prefer greatly the oxo tautomeric forms. In CCl4 solution an equilibrium between the free
C᎐O bond and a “dimeric” hydrogen-bonded form exists in which the latter predominates. Several IR and NMR (1H,
᎐
13C and 15N) spectral properties are shown to correlate with Hammett σ-values and/or atomic Mulliken charges and
bond orders, the latter being estimated by PM3 or AM1 semiempirical methods. The electron-impact mass spectra
were also recorded and the fragmentation mechanisms interpreted in terms of the energetics of the ionic species. In
addition, the geometric and electronic properties of 4-phenylthiazol-2(3H )-one (3a) and the related benzothiazol-
2(3H )-one (4) based on ab initio HF/6-31 G* calculations are compared with each other.
Hantzsch reaction of the corresponding α-bromoacetophen-
Introduction
ones with ammonium thiocarbamate in water at room temper-
Thiazole derivatives like thiamine pyrophosphate and penicillin
ature4 or the reaction of α-(methoxycarbonyl)sulfenyl ketones
form very important classes of naturally occurring com-
with ammonium acetate in acetic acid at 80 ЊC.5 The best syn-
pounds.1a Certain 2-acylimino-4-thiazolines show significant
thetic route to compounds 3 still starts from α-thiocyanato-
acetophenones (1), since this procedure is convenient and the
yields of the purified products are high.
activity as anticonvulsant agents1b and N-substituted benzo-
thiazolinones stimulate plant growth.1c Annelated compounds
like thiazolo[3,4-a]benzimidazoles possess reproducible in vitro
anti-HIV activity.1d We also found that 3,4-diarylthiazolin-2-
imines1e and dibromo derivatives of 4-substituted thiazol-
2(3H )-ones exhibited fungicidal acitivities.1f Dhami et al.1g pre-
pared several 4-arylthiazol-2(3H )-ones and found that some of
them inhibited the growth of a wide variety of bacteria at a
dilution of 1 : 1000.
Results and discussion
Synthesis
We prepared 3 by a modification of the method used by
Bariana et al.6 and postulated the reaction to occur via the
2-amino-5-aryl-1,3-oxathiolium salt (2).7,8
Although it has already been proved by means of spectro-
scopic methods and HMO calculations that in solution the
2-oxo form is the predominant tautomer of thiazolinones,1a
relatively little is known about the detailed electronic and struc-
tural characteristics of 4-arylthiazol-2(3H )-ones. Since they
offer an interesting model system for molecular design of fur-
ther biologically active compounds we decided to carry out a
detailed spectroscopic (IR, NMR, MS), correlative and theor-
etical investigation of several 4-(substituted-phenyl) derivatives.
The compounds studied (Scheme 1, 3a–p) were prepared by
cyclisation of α-thiocyanatoacetophenones (1) in acid solution.
The parent compound, 4-phenylthiazol-2(3H )-one (3a), was
obtained by Dyckerhoff2 and Arapides3 via the above pathway.
Other methods for preparation of 3 are known, such as the
IR Spectroscopy
Detailed infrared spectral data in dilute CCl4 solutions for 4-
arylthiazol-2(3H )-ones (3), except those4 for 3a, 3b, 3e and 3l,
have not been reported previously (Table 1). In concentrated
CHCl3 solutions the compounds exhibit a broad and rather
complex absorption band in the region of 1680–1635 cmϪ1
belonging to the stretching vibration of the C᎐O group. On the
᎐
other hand, in dilute CCl4 solutions this absorption band is
clearly split into two ν(C᎐O) maxima within the ranges of
᎐
1704–1695 cmϪ1 and 1661–1658 cmϪ1. This is in accord with the
results of Cornwell et al.,4 who suggested that substituted
thiazol-2(3H )-ones favour the oxo tautomer although in CCl4
there exists an equilibrium between the free molecules and
intermolecularly hydrogen bonded “dimeric” species (contain-
ing two hydrogen bonds, between the N–H of one molecule and
† Electronic supplementary information (ESI) available: NMR data,
including graphs; Cartesian coordinates for 3a and 4. See http://
www.rsc.org/suppdata/p2/b1/b106322g/
the C᎐O of the other and vice versa) based on variable concen-
᎐
DOI: 10.1039/b106322g
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336
This journal is © The Royal Society of Chemistry 2002
329

View Article Online
Table 1 Characteristic IR spectral data (in cmϪ1) for 4-arylthiazol-
2(3H )-ones 3
In CCl4
In CHCl3
Compound
ν(C᎐O)b
ν(C᎐O)f
ν(C᎐O)b
᎐
᎐
᎐
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
3p
1658.8
1660.3
1657.6
1661.2
1659.6
1667.6
1668.8
1659.8
1660.0
1658.8
1658.4
1657.6
1657.6
1635.2
1661.1
1678.6
1699.2a
1702.0a
1701.6
1703.6
1697.6a
1695.0
1697.6
1697.6
1698.8
1698.8
1698.4
1697.2a
1697.6
1697.8
1660.0a
1660.4a
1658.4
1660.8
1658.8a
1658.8
1659.6
1658.8
1659.6
1659.2
1659.2
1658.0a
1658.0
1658.5
b
b
—
—
c
c
—
—
a Values (in cmϪ1) given in ref. 4 for concentrations of 10Ϫ2 mol dmϪ3 in
CCl4: 3a, 1698, 1661; 3b, 1697, 1662; 3e, 1695, 1660; 3l, 1694, 1659.
b Insoluble in CCl4; ν(C᎐O) 1648 cmϪ1 in Nujol.4 c Insoluble in CCl4.
᎐
tration experiments, e.g. on 3a, 3b, 3e and 3l. The corresponding
ν(C᎐O) and ν(C᎐O) values in Table 1 are assigned to the vibra-
᎐
᎐
f
b
tions of the free and hydrogen bonded molecules in analogy
to Cornwell’s work. It is evident that in concentrated CHCl3
solutions the intermolecularly hydrogen bonded species (“dim-
eric” or those complexed with CHCl3) are preferred. The wave-
numbers for ν(C᎐O) vibrations in CCl solutions are always
᎐
b
4
lower, not very sensitive to substituents, and similar to those
observed for the solid state spectra of a series of 4Ј-[bis(4-aryl-
2-oxo-∆4-thiazolin-5-yl)methyl]benzo crown ethers,9 which
resemble compounds 3. On the contrary the wavenumbers of
Scheme 2 Numbering of atoms in 3a and benzothiazol-2(3H )-one (4).
The structures were derived from HF/6-31G* calculations.
ν(C᎐O) vibrations depend sufficiently on structural changes to
᎐
f
be applicable in correlation analysis.10 Attempts to obtain well
distinguished absorption bands for the ring ν(N–H) stretching
vibrations of the free species in dilute CCl4 solutions were not
successful, due to the very poor solubility of compounds 3 in
apolar, aprotic and non-interacting solvents.
substituent effects. Like H5, C5 is also most sensitive to sub-
stituent effects. The 15N chemical shifts are very insensitive to
substituent changes. The NH chemical shift of 3n deserves a
special mention since it is very broad and must be excluded
from a Σσ correlation. This is obviously due to some kind of
dynamic exchange between the NH and two OH protons in this
compound.
NMR Spectroscopy
For further structural characterisation of the studied com-
pounds their 1H (Table 2), 13C (Table 3) and 15N NMR chemical
shifts (Table 4, for atom numbering of the compounds see
Scheme 2) were recorded. In addition these data were used for
correlations10 with the calculated Mulliken charges and/or
Hammett Σσ values11 (see Correlation analysis). The H5 chem-
ical shift shows the largest range (ca. 1 ppm), i.e. strongest sub-
stituent effects, which is also reflected in its correlations. The C2
carbon shifts do not vary very much (<1 ppm; no correlations
either) but in general such a carbonyl carbon is fairly intact for
Mass spectrometry
According to our knowledge this is the first comprehensive
treatment of the mass spectra of 4-arylthiazol-2(3H )-ones.
Their Mϩ ions are very stable under EI, giving rise to the base
ؒ
peaks in all the spectra except that of the tert-butyl-substituted
compound 3k (Table 5). The two most important fragmentation
ؒ
pathways are the losses of HNCO and (CO ϩ CHS ) resulting
in the [M Ϫ HNCO]ϩ and ArCNHϩ ions, respectively. The
ؒ
Scheme 1 Synthesis and structures of 4-arylthiazol-2(3H )-ones 3a–p.
330
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336

View Article Online
Table 2 1H NMR chemical shifts (δ/ppm) of 4-arylthiazol-2(3H )-ones 3
Compd.
R
H-2Ј
H-3Ј
H-4Ј
H-5Ј
H-6Ј
H-5
NH
X
3a
3b
3c
3d
3e
3f
3g
3f
3i
3j
3k
3l
3m
3n
3o
3p
H
4Ј-Br
4Ј-Cl
3Ј,4Ј-Cl2
4Ј-CH3
2Ј,4Ј-(CH3)2
2Ј,5Ј-(CH3)2
3Ј,4Ј-(CH3)2
4Ј-C2H5
4Ј-CH(CH3)2
4Ј-C(CH3)3
4Ј-OCH3
4Ј-OC2H5
3Ј,4Ј-(OH)2
4Ј-NHCOCH3
4Ј-NO2
7.66
7.59
7.66
7.90
7.54
—
7.40
7.59
7.46
—
7.20
7.10
7.17
—
7.24
7.26
7.40
6.97
6.95
—
7.33
—
—
—
—
—
7.13
—
—
—
—
—
—
—
—
—
7.40
7.59
7.46
7.60
7.20
7.06
—
7.15
7.24
7.26
7.40
6.97
6.95
6.78
7.63
8.26
7.66
7.59
7.66
7.60
7.54
7.21
7.15
7.36
7.56
7.57
7.58
7.59
7.57
6.93
7.57
7.91
6.73
6.82
6.81
6.96
6.64
6.22
6.26
6.64
6.67
6.66
6.63
6.57
6.57
6.36
6.62
7.21
11.76
11.78
11.78
11.83
11.69
11.43
11.42
11.61
11.68
11.65
11.73
11.65
11.64
9.5p
—
—
—
—
2.29a
2.32b 2.30c
2.29d 2.29e
2.23f 2.21g
2.60h 1.17i
2.88j 1.19k
1.26l
—
7.45
7.56
7.57
7.58
7.59
7.57
7.03
7.57
7.91
3.77m
4.03n 1.32o
3.7p
7.63
8.26
11.70
12.05
10.08q 2.06r
—
a CH3. b 2Ј-CH3. c 4Ј-CH3. d 2Ј-CH3. e 5Ј-CH3 [d(H-2Ј) < d(H-5Ј)]. f 3Ј-CH3. g 4Ј-CH3. h CH2. i CH3. j CH. k (CH3)2. l C(CH3)3. m OCH3. n OCH2. o CH3.
p Broad signals. q NH. r COCH3.
Table 3 13C NMR chemical shifts (δ/ppm) of 4-arylthiazol-2(3H )-ones 3
Compd.
R
C-1Ј
C-2Ј
C-3Ј
C-4Ј
C-5Ј
C-6Ј
C-4
C-5
98.07
C-2
3a
H
4Ј-Br
4Ј-Cl
3Ј,4Ј-Cl2
4Ј-CH3
2Ј,4Ј-(CH3)2
2Ј,5Ј-(CH3)2
3Ј,4Ј-(CH3)2
4Ј-C2H5
4Ј-CH(CH3)2
4Ј-C(CH3)3
4Ј-OCH3
4Ј-OC2H5
3Ј,4Ј-(OH)2
4Ј-NHCOCH3
4Ј-NO2
129.75
128.81
128.48
129.97
127.03
127.79
130.38
127.29
127.25
127.38
127.02
122.40
122.25
121.66
124.45
135.35
124.93
126.82
126.58
126.58
124.81
135.84
132.86
125.90
124.89
124.92
124.67
126.34
126.32
112.86
125.48
125.76
128.82
131.69
128.79
131.82
129.34
131.24
130.48
136.63
128.14
126.67
125.51
114.20
114.61
145.64
119.03
124.17
128.50
121.62
133.04
130.86
138.06
138.33
129.66
136.80
144.31
148.89
151.09
159.43
158.70
146.35
139.62
146.70
128.82
131.69
128.79
130.86
129.34
126.51
134.93
129.77
128.14
126.67
125.51
114.20
114.61
115.90
119.03
124.17
124.93
126.82
126.58
124.81
124.81
129.04
129.45
122.27
124.89
124.92
124.67
126.34
126.32
116.91
125.48
125.76
133.98
132.79
132.74
131.50
133.99
133.92
133.98
134.05
133.97
133.97
133.93
133.74
133.75
134.59
133.83
132.06
173.05
172.79
172.81
172.63
173.03
172.63
172.54
172.94
172.98
172.98
173.02
173.05
173.04
173.43
173.14
172.57
3b
99.10
99.02
100.70
97.01
99.41
99.62
96.72
97.05
97.06
97.09
95.74
95.64
95.07
96.66
103.51
3c
3da
3eb
3fc
3gd
3he
3if
3jg
3kh
3li
3mj
3n
3ok
3p
a Assignments of C-3Ј and C-4Ј verified. b CH3 20.73 ppm. c 2Ј-CH3 19.94 and 4Ј-CH3 20.68 ppm. d 2Ј-CH3 19.48 and 5Ј-CH3 20.34 ppm. e 3Ј-CH3
19.32 and 4Ј-CH3 19.04 ppm. f CH2 27.84 and CH3 15.29 ppm. g CH 33.14 and (CH3)2 23.58 ppm. h C(CH3)3 34.30 and C(CH3)3 30.88 ppm. i OCH3
55.19 ppm. j OCH2 63.15 and CH3 14.51 ppm. k CO 168.60 and CH3 24.12 ppm.
Table 4 15N NMR chemical shifts (Ϫδ/ppm from ext. CH3NO2) of
4-arylthiazol-2(3H )-ones 3
ϩ
[M Ϫ HNCO]ϩ and ArCNH ions are often obscured by the
ؒ
substituent-specific fragmentations. Thus, peaks of [M Ϫ
HNCO]ϩ ions are absent from the spectrum of compound 3m
ؒ
Compound
NH
Other
(R = OEt), but there are prominent peaks of [M Ϫ C2H4 Ϫ
HNCO]ϩ ions. Similarly, the [M Ϫ C2H4 Ϫ COCHS] ions are
ϩ
ؒ
3a
230.3
230.9
230.7
231.2
230.4
225.2
225.0
230.4
230.3
229.5
230.3
231.2
—
—
—
—
—
—
—
—
three times more abundant than [M Ϫ COCHS]ϩ ions. The [M
Ϫ NO2 Ϫ COCHS]ϩ and [M Ϫ COCHS]ϩ ions are of compar-
able abundance in the spectrum of compound 3c. Formation of
3b
3c
3d
the abundant [M Ϫ COCH2]ϩ ions is followed by the usual
ؒ
3e
3f
HNCO and (CO ϩ CHS) losses in the case of compound 3d.
3g
The mechanism of the HNCO and (CO ϩ CHS) losses from
3h, 3i, 3j, 3k
the Mϩ ions in the absence of substituent-specific fragmenta-
ؒ
3l, 3m
3n
3o
3p
—
—
tions was deduced from quantum chemical calculations (PM3/
UHF calculations, MOPAC 6.0,12 see Computational details
below) of the reaction profiles for various single-bond cleavages
246.5 (NHCO)
11.5 (NO2)
in the Mϩ ions of compound 3a (R = H). The calculations
ؒ
demonstrate that the Mϩ ion of the oxo isomer A1 is thermo-
ؒ
metastable ion spectra show that the latter species are formed in
two steps via the less stable [M Ϫ CO]ϩ ions (ca. 5% RA).
dynamically by 68.6 kJ molϪ1 more stable than the alternative
ؒ
Apart from these main routes, there are numerous substituent-
hydroxy form B (Scheme 3). Moreover, the calculations predict
specific fragmentations of the Mϩ ions, such as losses of the
the existence of yet another cyclic form of the Mϩ ion, the
ؒ
ؒ
substituent (R = Me, Et, Cl, Br, NO2), methyl losses from the
four-membered C, which is less stable than the other two cyclic
i
t
Mϩ ions bearing branched alkyl substituents (R = Pr and Bu ),
and rearrangement-accompanied losses of C2H4 (R = OEt) and
CH2CO (R = NHAc). The substituent-specific losses are only
insignificant in the case of compounds 3l and 3n [R = OMe and
3,4-(OH)2]. The characteristic fragmentations leading to the
forms, but more stable than each of the open-chain forms A2
ؒ
and A3 (Scheme 3). The most stable Mϩ (A1) radical cation is
ؒ
practically flat, with both the phenyl and thiazole rings lying
in the same plane. The C4–N and C5–S bonds are strengthened
by aromatic conjugation, whereas the C2–N and C2–S bonds
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336
331

Table 5 Main fragments in the 70-eV EI mass spectra of 4-arylthiazol-2(3H )-ones (3). The base peaks correspond to the Mϩ ions unless specified otherwise
ؒ
[M Ϫ H]ϩ or
[M Ϫ Me]ϩ
[M Ϫ COCHS]ϩ
ϩ
Mϩ
177
[M Ϫ CO]ϩ
[M Ϫ HNCO]ϩ
(ArCNHϩ)
Arϩ (RC6H4
77(26)
)
[M Ϫ R]ϩ
[M Ϫ R Ϫ HNCO]ϩ
ؒ
ؒ
ؒ
Compd.
R
3a
3b
3c
3d
H
176(6)a
—
—
149(5)
148(9)c
148(4)c
—
134(19)
104(70)
-—
176(16)
176(16)
210(13), 212(5)
—
—
4Ј-Br
4Ј-Cl
3Ј,4Ј-Cl2
255(96) 257(100)
211(100) 213(37)
245(100) 247(67)
212(9) 214(8)
168(16) 170(6)
202(14) 204(9)
182(45) 184(45)
138(71) 140(22)
172(59) 174(38)
155(8) 157(8)
111(15) 113(5)
145(12) 147(8)
133(7)
167(6) 169(4)
—
175(12) [M Ϫ 2Cl]
b
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
3p
4Ј-CH3
191
205
205
205
205
219
233(90)
207
221
209
234
222
176(12)b
190(30)b
190(37)b
190(17)b
190(24)b
204(75)b
218(100)b
206(6)a, 192(6)b
—
163(3)
176(4)d
176(4)d
—
148(10)
162(6)e
162(6)f
162(9)g
162(3)
—
160(7)
164(8)h
—
118(54)
132(29)
132(25)
132(38)
132(32)
146(12)
—
134(44)
148(12), 120(38)*
136(45)
161(5), 119(38)*
149(26), 103(21)*
91(16)
—
—
105(5)
—
—
—
—
121(9)
—
92(6)*
76(13)*
147(12)j
b
b
b
2Ј,4Ј-(CH3)2
2Ј,5Ј-(CH3)2
3Ј,4Ј-(CH3)2
4Ј-C2H5
4Ј-CH(CH3)2
4Ј-C(CH3)3
4Ј-OCH3
4Ј-OC2H5
3Ј,4Ј-(OH)2
4Ј-NHCOCH3
4Ј-NO2
161(18)j
161(14)j
161(5)j
148(5)c
—
176(15)
147(15) [R = Me]
161(5) [R = Me]
—
149(16)
—
—
—
176(6)i
—
—
—
—
176(5)
193(35) [M Ϫ C2H4]
192(7)
192(64) [M Ϫ COCH2]
176(11)
208(4)a
—
—
181(3)
—
166(11)
149(17)*
—
148(6)c
—
a [M Ϫ H]ϩ. b [M Ϫ Me]ϩ. c [M Ϫ R Ϫ CO]ϩ. d [M Ϫ H Ϫ CO]ϩ. e Doublet comprising ca. 60% of [M Ϫ Me Ϫ CO]ϩ ions. f Doublet comprising ca. 70% of [M Ϫ Me Ϫ CO]ϩ ions. g Doublet comprising ca. 40% of [M Ϫ
Me Ϫ CO]ϩ ions. h Doublet comprising ca. 30% of [M Ϫ Me Ϫ CO]ϩ ions. i Doublet comprising ca. 20% of [M Ϫ Me Ϫ CO]ϩ ions. j [M Ϫ H Ϫ HNCO]ϩ. * Following the loss of either COCH2 (3o), C2H4 (3m), or NO2
ϩ
(3p) from the Mϩ . For instance: m/z 149 is [M Ϫ COCH2 Ϫ HNCO] in (3o), etc.
ؒ
ؒ

View Article Online
Table 6 Selected PM3 and AM1 semiempirical data (Mulliken charges and populations) for compounds 3a
Compound
p(C2᎐O6)/AM1
q(C4)/PM3
q(C5)/PM3
q(C1Ј)/PM3
q(C2Ј/6Ј)/AM1
q(N3)/AM1
q(C3Ј/5Ј)/AM1
᎐
3a
3b
3c
3d
3e
3f
3g
3h
3I
3j
3k
3l
3m
3n
3o
3p
1.8181
1.8224
1.8212
1.8234
1.8172
1.8161
1.8164
1.8165
1.8171
1.8172
1.8169
1.8174
1.8171
1.8163
1.8180
1.8321
Ϫ0.0555
Ϫ0.0635
Ϫ0.0591
Ϫ0.0635
Ϫ0.0518
Ϫ0.0514
Ϫ0.0541
Ϫ0.0513
Ϫ0.0516
Ϫ0.0522
Ϫ0.0514
Ϫ0.0447
Ϫ0.0437
Ϫ0.0510
Ϫ0.0531
Ϫ0.0874
Ϫ0.4674
Ϫ0.4622
Ϫ0.4645
Ϫ0.4612
Ϫ0.4695
Ϫ0.4677
Ϫ0.4669
Ϫ0.4697
Ϫ0.4696
Ϫ0.4693
Ϫ0.4696
Ϫ0.4721
Ϫ0.4726
Ϫ0.4711
Ϫ0.4664
Ϫ0.4465
Ϫ0.0508
Ϫ0.0427
Ϫ0.0514
Ϫ0.0353
Ϫ0.0597
Ϫ0.0694
Ϫ0.0546
Ϫ0.0543
Ϫ0.0597
Ϫ0.0571
Ϫ0.0583
Ϫ0.0946
Ϫ0.0972
Ϫ0.0453
Ϫ0.0666
Ϫ0.0196
Ϫ0.1664b
Ϫ0.4503
Ϫ0.4499
Ϫ0.4502
Ϫ0.4498
Ϫ0.4505
Ϫ0.4492
Ϫ0.4489
Ϫ0.4505
Ϫ0.4505
Ϫ0.4505
Ϫ0.4505
Ϫ0.4515
Ϫ0.4515
Ϫ0.4497
Ϫ0.4516
Ϫ0.4490
Ϫ0.1903b
Ϫ0.1683b
Ϫ0.1537b
Ϫ0.1588b
Ϫ0.1768b
Ϫ0.1488c/Ϫ0.1613d
Ϫ0.1610b
Ϫ0.0719c/Ϫ0.1687d
Ϫ0.1876b
Ϫ0.0578c/Ϫ0.1513d
Ϫ0.0680c/Ϫ0.1527d
Ϫ0.1567c/Ϫ0.1662d
Ϫ0.1626b
Ϫ0.1869c/Ϫ0.1947d
Ϫ0.1853c/Ϫ0.0956d
Ϫ0.0866c/Ϫ0.1822d
Ϫ0.1847b
Ϫ0.1635b
Ϫ0.1829b
Ϫ0.1640b
Ϫ0.1841b
Ϫ0.1294b
Ϫ0.2382b
Ϫ0.1286
Ϫ0.2406b
Ϫ0.1664c/Ϫ0.1623d
Ϫ0.1351b
ϩ0.0746c/Ϫ0.2262
Ϫ0.2312b
Ϫ0.1843b
Ϫ0.1226b
a Numbering of atoms, see Scheme 2. b Results from Boltzmann averaging (room temperature) of the calculated differences over the possible
orientations of the substituents; NMR shifts are the same due to rapid rotation. c In the case of compounds substituted at C3Ј, C3Ј and C5Ј are not
equivalent and show distinct NMR signals; therefore q-values for these atoms are listed separately (since the shifts are separate they are to be used as
separate entries in the correlations). d In the case of compounds substituted at C2Ј, C2Ј and C6Ј are not equivalent and show distinct NMR signals;
therefore q-values for these atoms are listed separately (since the shifts are separate they are to be used as separate entries in the correlations).
Scheme 3 Energy barriers for unimolecular decompositions (Ea, kJ molϪ1) approximated as differences between the calculated enthalpies of
formation (PM3/UHF) at 298 K for the transition states (TS) and the parent ion for 3a.
are weakened, which results in the facile ring opening and sub-
sequent loss of the CO and HNCO fragments (Scheme 3).
cleavage of the C2–S bond should result in the HNCO loss,
whereas the Mϩ fragmentation initiated by the C2–N bond
ؒ
An analysis of the isomeric Mϩ structures produced by the
cleavage is more likely to involve the consecutive losses of CO
ؒ
alternative bond cleavages in Mϩ (A1) suggests that the initial
and CHS (possibly via the four-membered cyclic isomer C).
ؒ
ؒ
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336
333

View Article Online
Table 7 Comparison of selected HF/6-31G* ab initio theoretical dataa and wavenumbers of C᎐O stretching vibrations for compound 3a and related
᎐
benzothiazol-2(3H )-one 4b
Data
Compound 3a
Compound 4
Total atomic charges
O6
C2
N3
C4
C5
S1
Ϫ0.5815
0.5790
Ϫ0.8262
0.3837
Ϫ0.5079
0.2453
O6
C2
N3
C3a
C7a
S1
Ϫ0.5751
0.5862
Ϫ0.8613
0.3962
Ϫ0.2715
0.2707
Overlap populations
C2–O6
C2–N3
N3–C4
C4–C5
C5–S1
S1–C2
0.5744
0.2450
0.2670
0.6428
0.2278
0.3491
C2–O6
C2–N3
N3–C3a
C3a–C7a
C7a–S1
S1–C2
0.5766
0.2483
0.2276
0.5508
0.2139
0.3382
Bond lengths/Å
C2–O6
C2–N3
N3–C4
C4–C5
C5–S1
S1–C2
1.207
1.374
1.398
1.328
1.751
1.781
C2–O6
C2–N3
N3–C3a
C3a–C7a
C7a–S1
S1–C2
1.205
1.373
1.390
1.385
1.759
1.786
Bond angles/Њ
S1–C2–N3
C2–N3–C4
N3–C4–C5
C4–C5–S1
C5–S1–C2
107.4
116.5
112.7
112.3
91.1
S1–C2–N3
108.5
116.8
112.2
111.1
91.3
C2–N3–C3a
N3–C3a–C7a
C3a–C7a–S1
C7a–S1–C2
Wavenumbers/cm؊1
Calculated
Experimental/CCl4
ν(C2᎐O6)
1763
1699.2
ν(C2᎐O6)
1772
1713.6c
᎐
᎐
ν(C2᎐O6)f
ν(C2᎐O6)f
᎐
᎐
a Only the data characterising the five-membered rings were selected. b For numbering of atoms see Scheme 2. c Taken from ref. 13.
Indeed, the calculated length of the C–N bond connecting the
In principle, for disubstituted derivatives two conformations
emerging HNCO fragment in the open-chain Mϩ (A2) ion is
as defined by the torsional angle N3–C4–C1Ј–C2Ј (see Scheme
2 for numbering of atoms) are possible. The energy differences
between these two conformations are very small. However, even
for derivatives unsubstituted at carbon atoms C2Ј and C6Ј or
C3Ј and C5Ј, respectively, the corresponding Mulliken charges
are different. Consequently, for the correlation analyses a
Boltzmann weighting on the calculated energy differences of
the possible conformations at room temperature was done for
the Mulliken charges.
ؒ
larger (1.46 Å) than that of the other C–N bond (1.35 Å) which
ϩ
should be cleaved to produce the [M Ϫ CO]ϩ ion from M
ؒ
ؒ
(A2). A transition state for the loss of HNCO from the Mϩ (A2)
structure was found to be 166 kJ molϪ1 higher in energy than
the parent ion. The search for a TS corresponding to the loss of
ؒ
CO from Mϩ (A2) did not converge to a saddle point. On the
ؒ
other hand, each of the C–S bonds connecting CO fragments in
ϩ
the Mϩ (A3) and M (C) forms is quite weak (1.84 and 1.93 Å,
ؒ
ؒ
respectively). The four-membered cyclic isomer of the [M Ϫ
For comparison Table 7 lists selected theoretical data based
on an HF/6-31G* ab initio approach and the calculated and
CO]ϩ ion is more stable than its open-chain counterpart
ؒ
(Scheme 3), which would easily cyclise into the former (the pre-
dicted energy of activation is only ca. 16 kJ molϪ1). The lowest-
energy fragmentation path thus leads to the cyclic isomer of [M
experimental wavenumbers for the C᎐O stretching vibration for
᎐
compound 3a and the related benzothiazol-2(3H )-one 4. Com-
pound 4 is a condensed aromatic analogue of thiazol-2(3H )-
one or its derivatives 3 and its spectral properties, interactions
with solvents and hydrogen bonding ability have been recently
thoroughly studied.13,14 The comparison shows that compound
4 exhibits a ν(C᎐O) absorption band which is 14.4 cmϪ1 higher
Ϫ CO]ϩ , although the subsequent loss of CHS occurs more
ؒ
ؒ
easily from the open form of [M Ϫ CO]ϩ (Scheme 3). The
ؒ
enthalpies of formation of the systems (PhC2HSϩ ϩ HNCO)
ؒ
ϩ
ؒ
and (PhCNH ϩ CHS) are similar, therefore the much greater
᎐
f
abundance of the [M Ϫ CO Ϫ CHS]ϩ ions compared to that of
than that of compound 3a. This is in good accord with the
calculated ab-initio C2–O6 overlap population of 4, which is
higher than the corresponding overlap population of 3a,
whereas the total negative charge at O6 as well as the corre-
sponding C2–O6 bond length for 4 are smaller than those for
3a. Based on the bond angles for 3a and 4 it is evident that no
significant difference is caused by the possible change in the
strain of the five-membered ring due to annelation. Finally the
overlap populations over the bonds of the five-membered
2(3H )-thiazole ring show that in the case of 3a the system is
significantly enriched by electrons, which are obviously well
conjugated with the carbonyl group which causes a decrease in
the ν(C᎐O) band position (∆ν = 14.4 cmϪ1) against the benzo-
[M Ϫ HNCO]ϩ in the spectrum of 3a (70% vs. 19% RA, Table
ؒ
5) can be explained by the higher energy of activation for the
HNCO cleavage from Mϩ (A2) (Scheme 3).
ؒ
Theoretical data
From the values obtained by semiempirical quantum chemical
methods, the Mulliken C᎐O bond orders of compounds 3
᎐
[p(C2᎐O6)] and the Mulliken charges at C1Ј, C2Ј/6Ј, C3Ј/5Ј, C4,
᎐
and C5 carbon atoms (q-values) derived by the PM3 method
[C1Ј, C4, and C5] and those derived by the AM1 method [C2Ј–
C6Ј] were selected as the most consistent for linear correlations
with spectral data.10 As to the Mulliken nitrogen atom charges
[q(N3)], the AM1 data appeared most suitable. These selected
semiempirical theoretical data are listed in Table 6.
᎐
f
thiazolinone analogue 4. This decrease is in good agree-
ment with the calculated difference of the ν(C᎐O) values (∆ν =
᎐
9 cmϪ1).
334
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336

View Article Online
Table 8 Correlation analysis (y = ax ϩ b) of selected spectral data for compounds 3a.a p refers to the Mulliken AM1 bond orders and q to the AM1
or PM3 charges
y
x
Nb
a
b
Rc
sd
pe
ν(C2᎐O6)f/CCl4
Σσ
p(C2᎐O6)/AM1
Σσ
Σσ
q(C2Ј/6Ј)/AM1
q(C2Ј/6Ј)/PM3
q(C3Ј/5Ј)/AM1
q(C3Ј/5Ј)/PM3
q(C3Ј/5Ј)/AM1
q(C3Ј/5Ј)/PM3
Σσ
13f
13f
13g
13
7.32 0.44
1699.5
221.1
7.62
7.40
138.7
136.8
141.9
142.2
0.981
0.964
0.914
0.926
0.843
0.840
0.830
0.817
0.890
0.896
0.863
0.865
0.958
0.989
0.739
0.967
0.914
0.805
0.813
Ϫ0.924
0.905
0.878
0.957
0.964
0.926
0.950
0.836
0.771
0.42
0.57
0.06
0.14
1.7
1.8
4.3
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
=0.00107
<0.0001
<0.0001
=0.00030
=0.00023
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
=0.00037
=0.00203
᎐
ν(C2᎐O6)f/CCl4
812.8 67.5
0.35 0.05
1.04 0.13
81.6 12.6
67.8 10.6
86.5 13.3
83.9 13.6
10.5 1.44
8.83 0.21
0.92 0.15
0.64 0.10
0.50 0.05
37.4 1.7
᎐
᎐
δH2Ј
δH3Ј
δ(13C2Ј/6Ј)
19g
19g
21
δ(13C3Ј/5Ј)
δH3Ј/5Ј
21
4.4
16h
16h
15
9.26
0.16
0.16
0.19
0.13
0.05
0.03
0.17
0.05
0.05
0.40
0.39
0.33
0.96
1.08
0.66
0.90
1.27
0.11
0.19
0.23
7.88
7.34
6.67
6.72
37.4
29.1
18.7
11.73
129.1
136.5
133.4
244.3
332.8
97.8
135.4
136.5
230.5
400.5
234.4
δH5Ј
δH5
Σσ
Σσ
16
13j
13j
16
δH5
q(C5)/AM1
q(C5)/AM1
q(C5)/PM3
Σσ
q(C4)/AM1
q(C4)/PM3
Σσ
q(C5)/AM1
q(C5)/PM3
Σσ
q(C1Ј)/AM1
q(C1Ј)/PM3
Σσ
42.0 10.2
25.6 2.0
13j
13j
15k
15k
16
δNH
0.30 0.04
62.5 12.8
51.5 10.2
Ϫ2.30 0.25
313.7 39.5
440.0 64.2
6.06 0.55
139.8 10.7
152.1 17.2
0.91 0.09
377.1 74.7
232.5 57.9
δ(13C4)
δ(13C4)
δ(13C5)
16
16
δ(13C5)
δ(13C1Ј)
15g
15g
13j
13j
13j
δ(15N3)
δ(15N3)
q(N3)/AM1
q(N3)/PM3
a For numbering of atoms see Scheme 2. b Number of compounds used in correlation (see also the last paragraph before computational details).
c Correlation coefficient. d Standard deviation. e p-value for the correlation. f Data for 3f excluded, no IR-data obtained for 3o and 3p (see Table 2).
g Data for 3n excluded. h Data for 3o excluded. j Data for 3f, 3g and 3n are excluded. k Data for 3d excluded.
Correlation analysis
emphasizing that the latter three compounds (3f, g, n) appear to
behave exceptionally since omitting them from several corre-
lations significantly improved the fits. For instance, in the case
of the H5 vs. q(C5)/AM1 correlation the plot with all shifts has
R = 0.739 whereas that excluding the above compounds has
R = 0.989 (Table 8). This was more or less in line with the
expectations.
One should note that usually the number of correlated
parameters is 16 (equal to the number of compounds; Table 8)
but less as stated above owing to some omissions or to the fact
that some of the positions are substituted (proton chemical
shifts). In the case of carbons the Mulliken AM1 charges are
different for C2Ј and C6Ј or C3Ј and C5Ј if some of these
positions are substituted. Therefore in these cases the number
of correlated parameters can be higher than 16 (up to 21).
The statistical correlations of IR and NMR spectral data with
Hammett σ constants11 of substituents R (or in fact with their
sums, Σσ for disubstituted compounds) and with some PM3
and AM1 theoretical data for compounds 3 are given in Table
8.10 The ν(C᎐O) wavenumbers exhibit significant linear correl-
᎐
f
ations with both Hammett σ values and Mulliken AM1 C᎐O
᎐
bond orders (p). The relatively high value of the slope (a =
7.32 cmϪ1) of ν(C᎐O) vs. Σσ correlation suggests a very efficient
᎐
f
transmission of electronic effects from the 4-aryl moiety to the
terminal carbonyl group attached to the 2(3H )-thiazole ring.
Satisfactory or good correlations can be found also between
the 13C NMR chemical shifts of C1Ј, C2Ј/6Ј, C3Ј/5Ј, C4, and C5
carbon atoms and the corresponding Mulliken AM1 and PM3
charges (q; Table 8). The relatively large value of the slope (a =
313.7 ppm/AM1 or 440.0 ppm/PM3) of the δ(13C5) vs. q(C5)
correlation indicates the high sensitivity of C5 to substituent
effects, which evidently plays a significant role in the trans-
Computational details
Molecular structures were calculated by the semiempirical
AM115 and PM316 Hamiltonians using the VAMP 4.40 pro-
gram package.17 Geometries were completely optimised by the
eigenvector (EF) following routine18 without any restriction
and using the PRECISE option. In addition, for compounds 3a
and 4 ab initio HF/6-31G* computations using the Gaussian 94
program suite19 were performed. Calculated frequencies are
scaled by a factor 0.8929.20
mission of electronic effects from the 4-aryl moiety to the C᎐O
᎐
group and indicates that it takes place via the C5 atom. Sim-
ilarly the C1Ј, C2Ј/6Ј, C3Ј/5Ј, and C4 carbon chemical shifts
show a good or at least a satisfactory correlation with the AM1
and PM3 Mulliken charges and those of C4 and C5 also with
the Σσ values (Table 8).
1
Several H chemical shifts (those of H2Ј, H3Ј, H5Ј, H5 and
NH) show a good or fair correlation with the Hammett con-
stants and in the case of H3Ј/5Ј and H5 also with the AM1 and
PM3 Mulliken charges of the respective carbon atoms (Table
8). Despite the fact that the 15N chemical shifts of the ring NH
of compounds 3a–p exhibit a very narrow range (Ϫ230.5
1.0 ppm), i.e. they show very weak substituent dependences, the
shifts (measured within 0.1 ppm) do correlate with both of the
above parameters (see below) excluding the 3,4-dihydroxy 3n
(due to intramolecular hydrogen bonding) and the o-methyl
substituted derivatives 3f and 3g, of which the latter two show a
Unimolecular decompositions of gas-phase ionic species
formed from 3a under EI were calculated by the PM3 method16
using the MOPAC 6.0 program12 with the PRECISE option
and the unrestricted Hartree–Fock (UHF) open-shell approx-
imation. Transition-state structures were located using either
TS or NLLSQ routines and characterised by force calculations
(FORCE routine) to verify that they had one, and only one,
negative force constant (imaginary frequency). The energy
barriers (Ea) for the unimolecular decompositions shown in
Scheme 3 correspond to differences between the calculated
standard enthalpies of formation (∆Hf298 K, MOPAC standard
very clear ortho-effect (δN = 225.1
0.1 ppm). It is worth
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336
335

View Article Online
output) of the transition state structures and respective parent
ions.
also evident that transmission of substituent effects to the C᎐O
group takes place via this atom. Finally, aided by semiempir-
᎐
ical molecular orbital calculations of the complete reaction
paths (transition states and intermediates), a detailed descrip-
tion of the mass spectrometric fragmentation behaviour of
4-arylthiazol-2(3H )-ones is deduced.
Experimental
Preparation of 4-arylthiazol-2(3H)-ones (3)7
The corresponding α-thiocyanatoacetophenone (0.1 mol) was
dissolved in glacial acetic acid (10 ml), 50% sulfuric acid (2 ml)
was added and then the mixture was heated to boiling for
10 min. The products 3 crystallised or had to be precipitated by
water. They were collected on suction and recrystallised from
ethanol–acetone.
Acknowledgements
The authors wish to thank The Finnish Academy (Grant no.
4284) and the Scientific Grant Agency of the Ministry of
Education of the Slovak Republic (Grant no. 1/4000/97) for
financial support.
Infrared spectra
References
The IR spectra were measured at room temperature in the
region of 1750–1600 cmϪ1 using a Zeiss Specord M 80 spec-
trometer. The measurements were carried out in CHCl3 and
CCl4 employing NaCl cells of 1 mm and 10 mm thickness.
CHCl3 and CCl4 were of spectral purity (Uvasol, Merck). The
concentration of the solution in CHCl3 was ca. 4 × 10Ϫ3 mol
dmϪ3. In CCl4 the concentrations were in the range between
10Ϫ4 and 4 × 10Ϫ4 mol dmϪ3. For measurements in CCl4 peak
positions were determined with an accuracy of 0.2 cmϪ1
against polystyrene standard spectra.
1 (a) A. Dondoni and P. Merino, in Comprehensive Heterocyclic
Chemistry II, Elsevier Science Ltd., 1st edn., 1996, vol. 3, pp.
373–474; (b) N. Ergenc and G. Capan, Farmaco, Ed. Sci., 1994, 49,
133; (c) V. Sutoris, A. Gaplovsky and V. Sekerka, Chem. Pap., 1993,
47, 260; (d ) A. Cimirri, S. Grasso, A.-M. Monforte, P. Monforte
and M. Zappala, Farmaco, Ed. Sci., 1991, 46, 817 and 925; (e)
H. Naumann, H. Dehne, C. Fieseler, K. Goetzschel, M. Pallas,
D. Schoenfelder, W. Mueller, W. Kochmann, K. Naumann and
W. Steinke, Ger. (East) DD 241 844/1985/1987 (Chem. Abstr., 1987,
107, 91 901); ( f ) H. Hoelzel, H. Dehne, C. Fieseler, W. Kochmann,
P. Krey, K. Naumann, K. Pede and M. Wind, Ger. (East) DD 200
402/1981/1983 (Chem. Abstr., 1983, 99, 153 858); (g) K. S. Dhami,
S. S. Arora and K. S. Narang, J. Org. Chem., 1962, 27, 2147.
2 R. Dyckerhoff, Ber. Dtsch. Chem. Ges., 1877, 10, 119.
3 L. Arapides, Justus Liebigs Ann. Chem., 1888, 249, 7.
4 S. P. Cornwell, P. T. Kaye, A. G. Kent and G. D. Meakins, J. Chem.
Soc., Perkin Trans. 1, 1981, 2340.
5 Y. Sanemitsu, S. Kawamura and Y. Tanabe, J. Org. Chem., 1992, 57,
1053.
6 D. S. Bariana, H. S. Sachdev and K. S. Narang, J. Indian Chem. Soc.,
1955, 32, 427.
7 J. Teller, Habilitation Thesis, University of Rostock, Germany, 1988,
pp. 41, 155.
8 J. Liebscher, in Houben-Weyl, Methoden der Organischen Chemie,
Thieme Verlag, Stuttgart–New York, 4th edn., 1994, vol. E 8b,
p. 153.
NMR Spectra
The 1H and 13C NMR spectra were run for 0.1–0.2 M DMSO-
d6 solutions at 30 ЊC with a Bruker Avance DRX500 or
DPX250 spectrometer working at 500.13 or 250.13 MHz for
proton and 125.77 or 62.89 MHz for carbon-13, respectively.
1H, 13C and 15N NMR chemical shift assignments are based
1
1
on PFG H, 13C HMQC,21,22 1H, 13C HMBC23 and H, 15N
HMBC experiments. The detailed lists of the acquisition and
processing parameters are available from E. K. on request.
Mass spectra
9 J. Teller, H.-J. Holdt and H. Dehne, Z. Chem., 1989, 29, 446.
10 For comparable correlative studies see: (a) A. Perjéssy, D. Rasala, D.
Loos and D. Piorun, Monatsh. Chem., 1997, 128, 871; (b) Hu Geng
Yuan, Ye Hua Hua, P. Koisˇ, N. Prónayová, Liu Quing and A.
Perjéssy, Heterocycl. Commun., 1998, 4, 39; (c) E. Kolehmainen, K.
Lappalainen, A. Perjéssy, M. Lácová and W. M. F. Fabian, Magn.
Reson. Chem., 1998, 36, 511; (d ) A. Perjéssy, K. Bowden, W. M. F.
The low-resolution EI mass spectra were obtained by using a
VG 7070E mass spectrometer (Manchester, UK) at 70 eV
(direct insertion probe, ion source temperature 160 ЊC). Elem-
ental compositions of fragment ions were determined using a
VG ZabSpec instrument within an average accuracy of ca. 3 ×
10Ϫ4 u based on accurate mass measurements at a resolution of
10 000–12 000 (10% valley definition) by the peak matching
technique, using perfluorokerosene (PFK) as a reference com-
pound. Metastable ion spectra (B/E and B2/E linked scan tech-
nique, decompositions in the 1FFR) were recorded with the VG
ZabSpec mass spectrometer.
ˇ
Fabian, O. Hritzová, N. Prónayová, Z. Susteková and A. Al. Najjar,
Monatsh. Chem., 1999, 130, 515 and references therein.
11 C. Hansch and A. Leo, Substituent Constants for Correlation
Analysis in Chemistry and Biology, Wiley, New York, 1979.
12 J. J. P. Stewart, MOPAC 6.0, QCPE No. 455, Indiana University,
Bloomington, IN.
13 A. Perjéssy and J. B. F. N. Engberts, Monatsh. Chem., 1995, 126, 871.
14 J. B. F. N. Engberts, G. R. Famini, A. Perjéssy and L. Y. Wilson,
J. Phys. Org. Chem., 1998, 11, 261.
15 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart,
J. Am. Chem. Soc., 1985, 107, 3902.
16 J. J. P. Stewart, J. Comput. Chem., 1989, 10, 209 and 221.
17 T. Clark, VAMP, Erlangen Vectorized Molecular Orbital Package,
Version 4.40, Computer-Chemie-Zentrum, University Erlangen-
Nürnberg, Germany, 1992.
Conclusions
The synthesis of a series of arylthiazolones with potential
biological activity was described. The molecular as well as
electronic structure of these compounds was characterised by
means of spectroscopic (IR, NMR, MS) and theoretical (semi-
empirical and ab initio) calculations. Infrared spectroscopy
shows that compounds 3 exist in solution mainly as lactam
tautomers, either in free or hydrogen-bonded dimeric forms.
Especially in concentrated CHCl3 solution the hydrogen-
bonded dimers and/or complexes with the solvent are the
dominant species. As indicated by good linear correlations with
18 J. Baker, J. Comput. Chem., 1986, 7, 385.
19 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johanson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Pepersson,
J. A. Montgomery, K. Raghavachari, M. A. Al-Lahm, V. G.
Zakrzewski, J. V. Ortiz, J. B. Foresman, C. Y. Peng, P. Y. Ayala, W.
Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L.
Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. J. P. Stewart,
M. Head-Gordon, C. Gonzales and J. A. Pople, Gaussian 94,
Revision B. 3, Gaussian, Inc., Pittsburgh, 1993.
Hammett σ values and AM1 C᎐O bond orders, the carbonyl
᎐
stretching frequency of the monomeric species is quite sensitive
to the effects of the substituents of the aryl ring. Transmission
of such effects appears to be rather efficient. Similarly, corre-
lations of Mulliken populations with the respective 1H
and/or 13C NMR chemical shifts indicate that the carbon atom
C5 is rather susceptible to effects of 4-aryl substituents. It is
20 J. A. Pople, A. P. Scott, M. W. Wong and L. Radom, Isr. J. Chem.,
1993, 33, 345.
21 A. Bax, R. H. Griffey and B. L. Hawkins, J. Magn. Reson., 1983, 55,
301.
22 A. Bax and S. Subramanian, J. Magn. Reson., 1986, 67, 565.
23 A. Bax and M. F. Summers, J. Am. Chem. Soc., 1986, 108, 2093.
336
J. Chem. Soc., Perkin Trans. 2, 2002, 329–336