Kinetics and mechanism of formation and decomposition of substituted 1-phenylpyrrolidin-2-ones in basic medium

Article abstract of DOI:10.1002/poc.468

A study of chemical behaviour of substituted 4-chloro-N-phenylbutanamides in aqueous solutions of sodium hydroxide showed that the substrate first undergoes ring closure to give substituted 1-phenylpyrrolidin-2-ones, which are subsequently hydrolysed to substitution derivatives of sodium 4-amino-N-phenylbutanoates. Kinetic measurements provided the values of dissociation constants pK_a and cyclization rate constants k_c in water at 25°C for 2-bromo-4-chloro-N-(4-nitrophenyl)butanamide [pK_a = 11.64 ± 0.01; k_c = (1.94 ± 0.03) × 10^-2 s^-1], 4-chloro-N-(4-nitrophenyl)butanamide [pK_a = 13.35 ± 0.02; k_c = (1.60 ± 0.02) × 10^-2 s^-1 and 4-chloro-2-methyl-N-(4-nitrophenyl)butanamide [pK_a = 13.55 ± 0.03; k_c = (7.61 ± 0.11) × 10^-2 s^-1]. The pK_a and k_c values of individual derivatives differ depending on the substitution at the α-position of the butanamide skeleton. In methanolic sodium methoxide solutions, the course of ring closure of 2-bromo-4-chloro-N-(4-nitrophenyl)butanamide is of similar nature but slower [K = 60.10 ± 0.08 and k_c = (6.52 ± 0.05) × 10^-3 s^-1]. The subsequent hydrolyses of substituted 1-phenylpyrrolidin-2-ones to substituted 4-aminobutanoic acids also have different courses with different derivatives and depend on the substituents in the aromatic and/or heterocyclic moiety. The rate-limiting step of hydrolysis of 1-(4-nitrophenyl)pyrrolidin-2-one consists of the non-catalysed decomposition of the tetrahedral intermediate. In the case of 3-bromo-1-(4-nitrophenyl)pyrrolidin-2-one at sodium hydroxide concentrations below 0.1 mol 1^-1, the rate-limiting step is the second reaction pathway, i.e. the hydroxide ion-catalysed decomposition of the tetrahedral intermediate. At sodium hydroxide concentrations above 0.1 mol 1^-1], the rate-limiting step shifts to formation of the tetrahedral intermediate. This formation of the intermediate is 140 times slower than its hydroxide ion-catalysed decomposition [k₃/k_-1 = (1.40 ± 0.04) × 10² 1 mol^-1]. Introduction of a 3-bromo substituent into 1-(4-nitrophenyl)pyrrolidin-2-one results in acceleration of all the reaction steps of hydrolysis and increases the acidity of the intermediate. Copyright

Full text of DOI:10.1002/poc.468

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2002; 15: 165–173
DOI:10.1002/poc.468
Kinetics and mechanism of formation and decomposition
of substituted 1-phenylpyrrolidin-2-ones in basic medium†
MilosÏ SedlaÂk,1* Ludmila HejtmaÂnkovaÂ,2 Pavla KasÏparovaÂ3 and JaromõÂr KavaÂlek1
1Department of Organic Chemistry, Faculty of Chemical Technology, University of Pardubice, 53210 Pardubice, Czech Republic
2
Ï
Research Institute for Pharmacy and Biochemistry, DolnõÂ Mecholupy 130, 10201 Prague, Czech Republic
3Department of Colloid Chemistry, Max-Planck-Institute of Colloids and Interfaces, Am MuÈ hlenberg, D-14424 Potsdam, Germany
Received 10 July 2001; revised 5 October 2001; accepted 16 October 2001
ABSTRACT: A study of chemical behaviour of substituted 4-chloro-N-phenylbutanamides in aqueous solutions of
sodium hydroxide showed that the substrate first undergoes ring closure to give substituted 1-phenylpyrrolidin-2-
ones, which are subsequently hydrolysed to substitution derivatives of sodium 4-amino-N-phenylbutanoates. Kinetic
measurements provided the values of dissociation constants pKa and cyclization rate constants kc in water at 25°C for
2-bromo-4-chloro-N-(4-nitrophenyl)butanamide [pKa = 11.64 Æ 0.01; kc = (1.94 Æ 0.03) Â 10À2 sÀ1], 4-chloro-N-(4-
nitrophenyl)butanamide [pKa = 13.35 Æ 0.02; kc = (1.60 Æ 0.02) Â 10À2
s
À1] and 4-chloro-2-methyl-N-(4-nitrophe-
nyl)butanamide [pKa = 13.55 Æ 0.03; kc = (7.61 Æ 0.11) Â 10À2 sÀ1]. The pKa and kc values of individual derivatives
differ depending on the substitution at the a-position of the butanamide skeleton. In methanolic sodium methoxide
solutions, the course of ring closure of 2-bromo-4-chloro-N-(4-nitrophenyl)butanamide is of similar nature but slower
[K = 60.10 Æ 0.08 and kc = (6.52 Æ 0.05) Â 10À3 sÀ1]. The subsequent hydrolyses of substituted 1-phenylpyrrolidin-
2-ones to substituted 4-aminobutanoic acids also have different courses with different derivatives and depend on the
substituents in the aromatic and/or heterocyclic moiety. The rate-limiting step of hydrolysis of 1-(4-
nitrophenyl)pyrrolidin-2-one consists of the non-catalysed decomposition of the tetrahedral intermediate. In the
case of 3-bromo-1-(4-nitrophenyl)pyrrolidin-2-one at sodium hydroxide concentrations below 0.1 mol lÀ1, the rate-
limiting step is the second reaction pathway, i.e. the hydroxide ion-catalysed decomposition of the tetrahedral
intermediate. At sodium hydroxide concentrations above 0.1 mol lÀ1, the rate-limiting step shifts to formation of the
tetrahedral intermediate. This formation of the intermediate is 140 times slower than its hydroxide ion-catalysed
decomposition [k3/kÀ1 = (1.40 Æ 0.04) Â 102 l molÀ1]. Introduction of a 3-bromo substituent into 1-(4-nitrophe-
nyl)pyrrolidin-2-one results in acceleration of all the reaction steps of hydrolysis and increases the acidity of the
intermediate. Copyright  2002 John Wiley & Sons, Ltd.
KEYWORDS: reaction kinetics; cyclization; solvolysis; 1-phenylpyrrolidin-2-ones; dissociation constants
INTRODUCTION
However, solvolysis of the primary g-lactam cycle3 can
take place subsequently, which is undesirable in the
synthesis of the cyclic system (Scheme 1).
Substituted 1-phenylpyrrolidin-2-ones represent impor-
tant building blocks in syntheses of a number of
pharmacologically active substances, such as lactamyl-
vinylcefalosporines1 and some other drugs used for the
prevention of arterial thrombosis.2 One of the possible
ways to synthesize the 1-phenylpyrrolidinone cycle
consists in intramolecular nucleophilic substitution of
chlorine in 4-chloro-N-phenylbutanamide derivatives by
the amide anion (SNi), which must be catalysed by bases.
The aim of this study was to carry out a kinetic
investigation of base-catalysed ring closure of 2-bromo-
4-chloro-N-(4-nitrophenyl)butanamide (1a), 4-chloro-N-
(4-nitrophenyl)butanamide (1b) and 2-methyl-4-chloro-
N-(4-nitrophenyl)butanamide (1c) in water and/or metha-
nol and to examine the subsequent solvolysis of the
cyclizates formed, namely 3-bromo-1-(4-nitrophenyl)-
pyrrolidin-2-one (2a), 1-(4-nitrophenyl)pyrrolidin-2-one
(2b), 3-methyl-1-(4-nitrophenyl)pyrrolidin-2-one (2c)
and 3-bromo-1-(4-cynophenyl)pyrrolidin-2-one (2d).
*Correspondence to: M. Sedla´k, Department of Organic Chemistry,
Faculty of Chemical Technology, University of Pardubice, 532 10
Pardubice, Czech Republic.
†
ˇ
ˇ
Dedicated to Professor Vojeslav Sterba on the occasion of his 80th
birthday.
EXPERIMENTAL
Contract/grant sponsor: Max-Planck Society.
Contract/grant sponsor: Grant Agency of the Czech Republic;
Contract/grant number: 203/01/0227.
2-Bromo-4-chloro-N-ꢀ4-nitrophenyl)butanamide ꢀ1a).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

´
166
M. SEDLAK ET AL.
Scheme 1
4-Nitroaniline (2.76 g; 25 mmol) was dissolved in a
mixture of dry acetone (25 ml) and triethylamine
(3.65 ml; 25 mmol). After cooling to 10°C, a solution
of 2-bromo-4-chlorobutanoyl bromide4 (6.6 g; 25 mmol)
in dry acetone (25 ml) was added drop by drop. The
reaction mixture was stirred at room temperature for 2 h.
The separated triethylammonium bromide was filtered
off and the filtrate was concentrated on a water-bath in
vacuum. The crystalline solid obtained was recrystallized
J = 9.3 Hz, 2H-arom); 8.25 (d, J = 9.3 Hz, 2H-arom);
10.64 (bs, 1H, NH). 13C NMR (ꢀ): 17.6 (CH3); 35.7
(CH2); 38.4 (CH); 43.3 (CH2Cl); 119.0 (CH-arom); 125.0
(CH-arom); 142.2 (C-arom); 145.5 (C-arom); 174.9
(C=O).
2-Bromo-4-chloro-N-ꢀ4-cyanophenyl)butanamide
1
ꢀ1d). 74%, m.p. 114–116°C. H NMR (ꢀ): 2.45 (m, 2H,
CH2); 3.82 (m, 2H, CH2Cl); 4.80 (t, J = 7.6 Hz, 1H,
CHBr); 7.83 (s, 4H-arom); 10.92 (bs, 1H, NH). 13C NMR
(ꢀ): 36.0 (CH2); 42.7 (CH2Cl); 46.4 (CHBr); 105.9 (C-
arom); 118.9 (CN); 119.6 (CH-arom); 133.4 (CH-arom);
1
(52%, m.p. 178–180°C). H NMR (ꢀ): 2.47 (m, 2H,
CH2); 3.78 (m, 1H, 1/2CH2); 3.88 (m, 1H, 1/2CH2); 4.83
(m, 1H, CHBr); 7.90 (d, J = 9.1 Hz, 2H-arom); 8.29 (d,
J = 9.1 Hz, 2H-arom); 11.11 (bs, 1H, NH). 13C NMR (ꢀ):
35.9 (CH2); 42.8 (CH2Cl); 46.7 (CHBr); 119.3 (CH-
arom); 125.1 (CH-arom); 142.9 (C-arom); 144.7 (C-
142.7 (C-arom); 167.1 (C=O).
3-Bromo-1-ꢀ4-nitrophenyl)pyrrolidin-2-one ꢀ2a). 2-
Bromo-4-chloro-N-(4-nitrophenyl)butanamide (1a) (1.6 g;
5 mmol) was dissolved in 50 ml of dichloromethane. The
solution was treated with a mixture of benzyltriethyl-
ammonium chloride (0.01 g) and sodium hydroxide
solution (50%; 0.4 g; 5 mmol) added at room tempera-
ture. The reaction mixture was stirred vigorously for 2 h.
The organic phase was separated, dried with anhydrous
sodium sulphate and filtered with charcoal (0.1 g). The
filtrate was concentrated in a vacuum evaporator to one-
third of its original volume and then added to hexane
(400 ml) with vigorous stirring. The precipitated crystal-
line solid was collected by suction and dried in a
vacuum desiccator (70%, m.p. 98–100°C). 1H NMR (ꢀ):
2.42 (m, 1H, 1/2CH2); 2.84 (m, 1H, 1/2CH2); 4.04 (m,
2H, CH2N); 5.00 (m, 1H, CHBr); 8.02 (d, J = 9.2 Hz, 2H-
arom); 8.32 (d, J = 9.2 Hz, 2H-arom). 13C NMR (ꢀ):
29.90(CH2); 46.4 (CHBr); 46.6 (CH2N); 119.4 (CH-
arom); 124.7 (CH-arom); 143.2 (C-arom); 144.6 (C-
arom); 167.4 (C=O).
Derivatives 1b±d. These were synthesized in the same
way from 4-nitroaniline, 4-aminobenzonitrile, 4-chloro-
butanoyl chloride and 2-methyl-4-chlorobutanoyl chlor-
ide,5 respectively.
4-Chloro-N-ꢀ4-nitrophenyl)butanamide ꢀ1b). 64%, m.p.
1
101–102°C. H NMR (ꢀ): 2.10 (m, 2H, CH2); 2.60 (t,
J = 7.3 Hz, 2H, CH2); 3.76 (t, J = 6.5 Hz, 2H, CH2Cl);
7.88 (d, J = 8.8 Hz, 2H-arom); 8.25 (d, J = 8.8 Hz, 2H-
arom); 10.64 (bs, 1H, NH). 13C NMR (ꢀ): 27.7 (CH2);
39.8 (CH2); 45.2 (CH2Cl); 118.8 (CH-arom); 125.1(CH-
arom); 142.1 (C-arom); 145.4 (C-arom); 171.4 (C=O).
4-Chloro-2-methyl-N-ꢀ4-nitrophenyl)butanamide ꢀ1c).
58%, m.p. 121–123°C. 1H NMR (ꢀ): 1.21 (d, J = 6.9 Hz,
3H, CH3); 1.88 (m, 1H, 1/2CH2); 2.15 (m, 1H, 1/2CH2);
2.80 (m, 1H, CH); 3.69 (m, 2H, CH2Cl); 7.90 (d,
arom); 170.5 (C=O).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

FORMATION AND DECOMPOSITION OF 1-PHENYLPYRROLIDIN-2-ONES
1
167
Substituted 1-phenylpyrrolidin-2-ones 2b±d. These
were synthesized by the same method from the
corresponding amides 1b–d.
83%, m.p. 121–123°C. H NMR (ꢀ): 1.16 (d, J = 7.0 Hz,
3H, CH3); 1.64 (m, 1H, 1/2CH2); 1.93 (m, 1H, 1/2CH2);
2.51 (m, 1H, CH); 3.20 (m, 2H, CH2); 6.67 (d, J = 8.0 Hz,
2H-arom); 7.37 (bt, J = 5.0 Hz, 1H, NH); 8.02 (d,
J = 8.0 Hz, 2H-arom). 13C NMR (ꢀ): 17.1 (CH3); 32.1
(CH2); 36.5 (CH); 40.5 (CH2N); 110.9 (CH-arom); 126.4
(CH-arom); 135.7 (C-arom); 154.6 (C-arom); 177.3
1-ꢀ4-Nitrophenyl)pyrrolidin-2-one ꢀ2b). 65%, m.p. 127–
129°C. 1H NMR (ꢀ): 2.13 (m, 2H, CH2); 2.61 (t,
J = 8.0 Hz, 2H, CH2); 3.93 (t, J = 7.0 Hz, 2H, CH2 N);
7.95 (d, J = 8.3 Hz, 2H-arom); 8.26 (d, J = 8.3 Hz, 2H-
arom). 13C NMR (ꢀ): 17.3 (CH2); 32.5 (CH2); 48.1
(CH2N); 118.7 (CH-arom); 124.7 (CH-arom); 142.4 (C-
(C=O).
4-Amino-2-bromo-N-ꢀ4-cyanofenyl)butanoic acid ꢀ3d).
72%, m.p. 130–132°C. 1H NMR (ꢀ): 2.11 (m, 1H,
1/2CH2); 2.28 (m, 1H, 1/2CH2); 3.23 (m, 2H, CH2N);
4.55 (m, 1H, CHBr); 6.66 (d, J = 7.2 Hz, 2H-arom); 6.77
arom); 145.3 (C-arom); 175.2 (C=O).
3-Methyl-1-ꢀ4-nitrophenyl)pyrrolidin-2-one ꢀ2c). 73%,
1
(
bs, 1H, NH); 7.47 (d, J = 7.2 Hz, 2H-arom). 13C NMR (ꢀ):
m.p. 179–180°C. H NMR (ꢀ): 1.22 (d, J = 7.1 Hz, 3H,
CH3); 1.77 (m, 1H, 1/2CH2); 2.38 (m, 1H, 1/2CH2); 2.76
(m, 1H, CH); 3.90 (m, 2H, CH2N); 8.00 (d, J = 9.2 Hz,
2H-arom); 8.30 (d, J = 9.2 Hz, 2H-arom). 13C NMR (ꢀ):
15.8 (CH3); 26.2 (CH2); 37.9 (CH); 46.1 (CH2N); 118.7
(CH-arom); 124.7 (CH-arom); 142.4 (C-arom); 145.5 (C-
33.5 (CH2); 39.8 (CH2N); 45.1 (CHBr); 96.0 (C-arom);
111.8 (CH-arom); 120.5 (CN); 133.4 (CH-arom); 151.9
(C-arom); 170.4 (C=O).
Methyl 4-amino-2-bromo-N-ꢀ4-nitrophenyl)butanoate
ꢀ4a). A suspension of 4-amino-2-bromo-N-(4-nitrophe-
nyl)butanoic acid (3a) (1g; 3.3 mmol) in diethyl ether
(30 ml) was treated with a freshly prepared solution of
diazomethane (0.21 g; 5 mmol) in diethyl ether (30 ml).
The suspension was stirred overnight whereby it
gradually changed into solution. The solution was
evaporated on a water-bath and the evaporation residue
was recrystallized from methanol (81%, m.p. 95–97°C).
1H NMR (ꢀ): 2.18 (m, 1H, 1/2CH2); 2.38 (m, 1H,
1/2CH2); 3.33 (m, 2H, CH2N); 3.75 (s, 3H, CH3); 4.71
(m, 1H, CHBr); 6.68 (d, J = 9.3 Hz, 2H-arom); 7.39
(bs, 1H, NH); 8.03 (d, J = 9.3 Hz, 2H-arom). MS (m/z,%):
arom); 177.5 (C=O).
3-Bromo-1-ꢀ4-cyanophenyl)pyrrolidin-2-one ꢀ2d). 64%,
1
m.p. 112–114°C. H NMR (ꢀ): 2.41 (m, 1H, 1/2CH2);
2.81 (m, 1H, 1/2CH2); 3.99 (m, 2H, CH2N); 4.98 (m, 1H,
CHBr); 7.95 (m, 4H-arom). 13C NMR (ꢀ): 29.0 (CH2);
46.2 (CH2N); 46.7 (CHBr); 106.6 (C-arom); 118.9 (CN);
119.6 (CH-arom); 133.2 (CH-arom); 142.8 (C-arom);
170.2 (C=O).
4-Amino-2-bromo-N-ꢀ4-nitrophenyl)butanoic acid ꢀ3a).
2-Bromo-4-chloro-N-(4-nitrophenyl)butanamide (1a) (1.2 g;
3.7 mmol) was mixed with water (10 ml) and aqueous
sodium hydroxide solution (50%; 4 g; 5 mmol). After
stirring for 4 h at room temperature, the mixture was
acidified with dilute hydrochloric acid (pH 7). The
crystalline solid that precipitated on cooling was collected
by suction on a sintered-glass filter, washed with a small
amount of water and recrystallized from water (92%, m.p.
126–128°C). 1H NMR (ꢀ): 2.15 (m, 1H, 1/2CH2); 2.35 (m,
1H, 1/2CH2); 3.33 (m, 2H, CH2N); 4.58 (m, 1H, CHBr);
6.69 (d, J = 9.0 Hz, 2H-arom); 7.39 (bs, 1H, NH); 8.04 (d,
J = 9.0 Hz, 2H-arom). 13C NMR (ꢀ): 33.5 (CH2); 39.8
(CH2N); 45.1 (CHBr); 110.9 (CH-arom); 126.3 (CH-
‡
‡
[M ‡ H] 317.1, 97; [M ‡ H] with 81Br 319.1, 100;
‡
‡
[M ‡ H À HBr] 237.1, 28; [M ‡ H À HBr À NO]
207.1, 19.
Methyl 4-amino-N-ꢀ4-nitrophenyl)butanoate ꢀ4b). This
was prepared similarly from acid 3b (83%, m.p. 150–
1
152°C). H NMR (ꢀ): 1.95 (m, 2H, CH2); 2.44 (m, 2H,
CH2); 3.24 (m, 2H, CH2N); 3.66 (s, 3H, CH3); 6.24 (bs,
1H, NH); 6.57 (m, 2H-arom); 8.00 (m, 2H-arom). 13C
NMR (ꢀ): 23.3 (CH2); 30.5 (CH2); 41.6 (CH2N); 50.6
(CH3); 110.1 (CH-arom); 125.4 (CH-arom); 136.0 (C-
arom); 153.5 (C-arom); 172.5 (C=O).
The results of elemental analyses of the individual
compounds agreed with the calculated values.
arom); 136.1 (C-arom); 154.3 (C-arom); 170.5 (C=O).
Acids 3b±d. These were synthesized in the same way
from the corresponding amides 1b–d.
1
Measurement of NMR spectra. The H and 13C NMR
4-Amino-N-ꢀ4-nitrophenyl)butanoic acid ꢀ3b). 88%,
m.p. 181–183°C. 1H NMR (ꢀ): 1.83 (m, 2H, CH2);
2.38 (t, J = 7.2 Hz, 2H, CH2); 3.22 (m, 2H, CH2N); 6.67
(d, J = 9.9 Hz, 2H-arom); 7.37 (bs, 1H, NH); 8.03 (d,
J = 9.9 Hz, 2H-arom). 13C NMR (ꢀ): 23.8 (CH2); 31.1
(CH2N); 41.6 (CH2); 110.8 (CH-arom); 126.3 (CH-
spectra were measured at 360.14 and 90.57 MHz,
respectively, on a Bruker AMX 360 apparatus. For the
measurements, the substances were dissolved in hexa-
deuteriodimethyl sulphoxide (DMSO-d6). The chemical
shifts ꢀH and ꢀC were referenced to the middle signal of
multiplet of the solvent (ꢀH = 2.55 ppm; ꢀC = 39.6 ppm).
The CH, CH2, CH3 and C groups in the 13C NMR spectra
were differentiated by the APT method. In the case of 4a,
only the 1H NMR spectrum is given, because the
arom); 135.7 (C-arom); 154.6 (C-arom); 174.5 (C=O).
4-Amino-2-methyl-N-ꢀ4-nitrophenyl)butanoic acid ꢀ3c).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

´
M. SEDLAK ET AL.
168
substance decomposed during the 13C NMR measure-
ment. Therefore, we present here also the mass spectrum.
The mass spectrum was measured with a VG Platform
II mass spectrometer (Micromass, Manchester, UK)
using atmospheric pressure chemical ionization (APCI)
and a quadrupole analyser (0–3000 Da). Before the mass
spectrometer there was inserted a separation apparatus
consisting of a Model 616 high-pressure pump, Model
717 autosampler and Model 966 UV detector (all from
Waters, Milford, MA, USA), and a Separon SGX C18
octadecylsilica glass cartridge column (150 Â 3 mm i.d.,
7 mm particle size) (Tessek, Prague, Czech Republic). A
mixture of 70% acetonitrile and 30% redistilled water
was used as the mobile phase. The effluent from the
liquid chromatography was introduced directly into a
quadrupole mass spectrometer equipped with APCI
probe operated in the positive ion mode. The data were
acquired in the m/z range 15–600 at 1.9 s per scan. In the
APCI mode, the temperatures of the ion source and of the
probe were 100 and 500°C, respectively. The cone
voltage was set at 10 V for positive-ion APCl.
acids 3a–d were isolated on a preparative scale. Under
both the preparative and the kinetic (pseudo-first order
reaction) reaction conditions, the a-bromo derivatives 1a
and 1d unambiguously gave the g-lactam cycle, the a-
lactams not being formed.6 Also, we never observed the
alternative ring closure reaction involving the oxygen
anion as internal nucleophile and giving imino lactones.7
This finding agrees with earlier papers7,8 reporting the
formation of imino lactones from 4-halo-N-substituted
butyramides in aqueous media at pH values below 10.
The observed rate constants of ring closure were
measured at the isosbestic points of hydrolysis (Table 1).
From the dependence of observed rate constant kobs on
sodium hydroxide concentration (Fig. 1) it can be seen
that with derivative 1a at lower sodium hydroxide
concentrations the first reaction (ring closure, formation
of 2a) is faster. At higher concentrations of sodium
hydroxide (>0.025 mol lÀ1), however, the second reac-
tion (hydrolysis, formation of 3a) is accelerated and
becomes faster than the ring closure.
First, we studied in more detail the cyclization
reactions of amides 1a–c to 2a–c, which differ from
each other by substitution at the a-position of the
butanamide skeleton. The course of ring closure of N-
phenylbutanamides can be expressed by Scheme 2.
The formation of amide anion is a fast pre-equilibrium
with a subsequent rate-limiting step, namely the intra-
molecular nucleophilic substitution of a chlorine atom
with concomitant formation of a five-membered ring.
The rate of the cyclization reaction followed under the
conditions of a pseudo-first order reaction can be
Kinetic measurements. The spectrophotometric
measurements were carried out on an HP UV/VIS 8453
diode-array apparatus in a 1 cm quartz cell with a lid at
the temperatures specified in Table 1. Before the
measurements proper, the time dependence of the
spectral change was measured for hydrolysis and
methanolysis of substituted 1-phenylpyrrolidin-2-ones
2a–d in the interval 200–1000 nm. These preliminary
experiments revealed isosbestic points and showed
suitable wavelengths for kinetic measurements. The
cyclization reactions of substituted butanamides 1a–c
(Table 1) were followed at the wavelengths of the
isosbestic points of solvolyses. The cell was charged with
2 ml of sodium hydroxide or sodium methoxide solution.
After attaining the chosen temperature, 20 ml of metha-
nolic solution of substrate was injected into the cell, so
the resulting concentration of substrate in the cell was
about 1 Â 10À4 mol lÀ1. The measured time dependence
of absorbance was treated by an optimization program to
obtain the pseudo-first-order rate constants kobs
.
RESULTS AND DISCUSSION
The acylation of substituted anilines with 4-chlorobutyryl
chloride, 2-bromo-4-chlorobutyryl chloride or 4-chloro-
2-methybutyryl chloride gave the respective substituted
4-chloro-N-phenylbutanamides 1a–d. In accordance with
what had been suggested, the amides 1a–d underwent
consecutive reactions in sodium hydroxide solution, the
first step being ring closure (i.e. formation of 2a–d) and
the next step the hydrolysis of primary substituted 1-
phenylpyrrolidin-2-ones to substituted sodium 4-amino-
butanoates 3a–d (Scheme 1). The substituted 1-phenyl-
pyrrolidin-2-ones 2a–d and substituted 4-aminobutanoic
Figure 1. Dependence of observed rate constant 2kobs) on
concentration of sodium hydroxide 2cNaOH) for cyclization
reaction 1a → 2a 2*) and hydrolysis reaction 2a → 3a 2*) at
25°C
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

FORMATION AND DECOMPOSITION OF 1-PHENYLPYRROLIDIN-2-ONES
169
Scheme 2
expressed by Eqn. (1), derived from Scheme 2:
concentration, the observed rate constant kobs is approxi-
mately comparable to the observed rate constant of
methyl derivative 1c, and about 5.5 times higher than that
of the unsubstituted derivative (1b). The initial consider-
able increase in the observed rate constant (kobs) for 1a is
caused by increased acidity of the amide nitrogen atom
due to the bromine substituent. The cyclization rate
constants kc (25°C) of derivatives 1a and 1b (Table 1) are
very similar, but kc for 1c is more than three times higher.
In the derivatives carrying a bromine substituent (1a) or
methyl group (1c) at the a-position of the respective
butanamide, the cyclization rate constants kc can be
affected in the following ways:
v ˆ kobscS ˆ kc‰NÀŠ
ꢀ1†
where cS is the total concentration of substrate, [NÀ] the
actual concentration of the anion and kc the rate constant
of ring closure. From the experimentally found depen-
dences of the observed rate constant (kobs, sÀ1) on sodium
hydroxide concentration (cNaOH, mol lÀ1) (Fig. 2), it
follows that increasing concentration of hydroxide ion
causes a non-linear increase in the observed rate constant
to the same extent as the concentration [NÀ] of anion is
increased. In the limiting case, when all the substrate has
been converted into the respective anion, the rate
becomes independent of the concentration of hydroxide
ion (the slope of the mentioned dependence approaches
zero).
1. favourable entropy effects: bulky substituents present
in the chain undergoing the ring closure can increase
The dependence depicted in Fig. 2 is expressed by
kcK‰OHÀŠ
kobs
ˆ
ꢀ2†
1 ‡ K‰OHÀŠ
The experimental values and Eqn. (2) were treated by the
optimization program to give the values of the constants
K and kc (Scheme 2). The experimental points were fitted
by a curve corresponding to Eqn. (2) into which the
calculated values of both K and kc (Fig. 2) had been
introduced. The values of the constants for the hydroxide
ion-catalysed ring closure of amides 1a–c to correspond-
ing lactams 2a–c are summarized in Table 1, where
pKa = pKw À logK, pKw being the ionic product of water.
From the values presented in Table 1, it follows that
substrate 1a (having a bromine substituent at the a-
position) is two orders of magnitude more acidic than the
unsubstituted derivative (1b). This difference corre-
sponds to the difference between the pKa values of acetic
and bromoacetic acids.9 Derivatives 1b and 1c differ in
their pKa values only slightly, the same being true for
propanoic and 2-methylpropanoic acids.10 From Fig. 2, it
can be seen that with derivative 1a the maximum value
(kobs) is already reached at a sodium hydroxide
concentration of 0.1 mol lÀ1. At this sodium hydroxide
Figure 2. Dependence of observed rate constant 2kobs) on
concentration of sodium hydroxide 2cNaOH) for cyclization
reactions 1a → 2a 2*), 1b → 2b 2*) and 1c → 2c 2&) at
25°C. The experimental points are interlaced by curves
corresponding to Eqn. 22)
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

´
M. SEDLAK ET AL.
170
Table 1. Constants of cyclisation 2kc, K), pKa and values of activation parameters for cyclization reaction of butanamides 1a±c in
solution of sodium hydroxide
Compound
Temperature (°C) 102 kc (sÀ1
)
K
pKa
DH≠ (kJ molÀ1
)
DS≠ (J molÀ1 KÀ1
)
1a
1b
25
15
20
25
32
40
15
20
25
32
40
1.94 Æ 0.03
229 Æ 0.3
11.64 Æ 0.01
0.53 Æ 0.01
0.99 Æ 0.01
1.60 Æ 0.02
3.70 Æ 0.03
7.01 Æ 0.02
2.24 Æ 0.01
4.44 Æ 0.05
7.61 Æ 0.11
13.3 Æ 0.2
28.9 Æ 0.1
6.71 Æ 0.10
5.10 Æ 0.12
4.43 Æ 0.10
3.27 Æ 0.15
3.16 Æ 0.13
2.65 Æ 0.41
2.95 Æ 0.36
2.83 Æ 0.12
4.07 Æ 0.36
4.05 Æ 0.45
13.35 Æ 0.02
74 Æ 3
36 Æ 6
1c
13.55 Æ 0.03
72 Æ 3
21 Æ 6
the kc values by as much as several orders of
magnitude;11
2. the negative inductive field effect (ÀI) of the bromine
substituent lowers the nucleophilicity of reagent and
hence also the kc value;
3. the positive inductive field effect (‡I) of the methyl
group increases the nucleophilicity of the reagent and
hence also the kc value.
and the values of the cyclization rate constant cannot be
separately determined by the method adopted by us.
We also studied the second reaction step, i.e.
hydrolysis (opening of the pyrrolidine cycle), with
derivatives 2a–d (Scheme 3) (the corresponding amides
1a–d were also used as the substrates). The hydrolysis
under both preparative conditions and conditions of
pseudo-first-order kinetics at 25°C gave sodium salts of
substituted 4-amino-N-phenylbutanoic acids 3a–d as the
only products. The alkaline hydrolysis of amides and
lactams has been studied extensively.14,15 Tertiary
amides and secondary lactams show simpler behaviour
than others as concomitant ionization of the weakly
acidic amide or lactam NH cannot occur. Evidence exists
for both first- and second-order reactions in analogous
hydroxide anion reactions.14 A simplified and general-
ized mechanistic pathway for the alkaline hydrolysis of
secondary lactams, where ionization of the substrates is
not possible, is shown in Scheme 3. Moreau et al.16 have
reviewed their studies on various series of amides,
including some N-methylformanilides and N-methylace-
tanilides in which the latter amides conform to the pattern
shown in Scheme 3. The alkaline hydrolysis of a series of
substituted N-methylformanilides,17 N-aryl-b-, -g- and
-ꢀ-lactams3 in water and aqueous DMSO have been
investigated. An important study of the alkaline hydro-
lysis of N-aryl-b-lactams and N-methylacetanilides in
water was made by Blackburn and Plackett.18 The results
were in contrast with those for the corresponding g- and
ꢀ-lactams and N-methylacetanilides. The Hammett r
value for substitution in the N-aryl-b-lactams clearly
indicated rate-determinig attack of hydroxide ion on the
b-lactam ring, followed by rapid ring fission. The other
lactams and acetanilides appear to have as the rate-
limiting step the hydroxide ion-catalysed decomposition
of a tetrahedral intermediate. All systems showed a first-
order dependence on hydroxide ion concentration. From
our experimental dependences of the logarithms of the
observed rate constants of hydrolysis on the logarithms of
sodium hydroxide concentration, it was possible to
determine the reaction order in OHÀ ion. The slopes of
On comparing the kc values of 1a and 1b, one can see
that they are almost the same, i.e. the above effects 1 and
2 on the cyclization rate must be similar. The kc value of
methyl derivative 1c is increased (compared with 1a and
1b) for the reasons 1 and 3 above. In order to be able to
compare quantitatively the entropy advantage for cycli-
zation due to the substitution in the side-chain, the
activation parameters of the cyclization reaction (kc) of
derivatives 1b and 1c were experimentally determined
(Table 1). The values found for the activation enthalpies
DH‡ are almost identical. The difference between the
found activation entropy values DS‡ of 1b and 1c is
indistinct and close to the experimental error. The
magnitude of difference between the two DS‡ values is
similar to that of acid-catalysed cyclizations of sub-
stituted N-phenylbutanoic acids to substituted 1-phenyl-
pyrrolidin-2-ones.12
Furthermore, we studied the ring closure reaction
1a → 2a in methanolic solutions of sodium methoxide,
and found that the dependence of the observed rate
constant on sodium methoxide concentration has a
similar character to that for the ring closure in aqueous
solutions of sodium hydroxide. The values found from
the measured dependence are kc = (6.52 Æ 0.05) Â 10À3
sÀ1 and K = 60.10 Æ 0.08 at 25°C. The decrease in the
cyclization rate 1a → 2a (kc) and also the decrease in the
equilibrium constant of formation of the reactive anion
can be explained, first of all, by changes in the solvation
ability on going from water to methanol.13 The rate of
cyclization 1b → 2b in methanolic solutions of sodium
methoxide is comparable to the rate of methanolysis of
2b to methyl 4-amino-N-(4-nitrophenyl)butanoate (4b),
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

FORMATION AND DECOMPOSITION OF 1-PHENYLPYRROLIDIN-2-ONES
171
Scheme 3
these dependences have different values for the indivi-
dual derivatives 2a–d differing in substitution of the
aromatic and/or heterocyclic moiety. The hydroxide ion-
catalysed hydrolysis of the substituted N-phenyl-2-
pyrrolidinone ring can be interpreted by Scheme 3.
With 2b and 2c the slopes of dependences of logkobs vs
logcNaOH are unity, which means that the main reaction
path is the non-catalysed decomposition of intermediate
InÀ (Scheme 3). This can be simplified by the kinetic
Scheme 4.
density at the carbonyl carbon of the amide group. The
methanolysis 2b → 4b in methanolic sodium methoxide
has the same character as the hydrolysis but is about
In this case the observed rate constant can be expressed
by
k1k2
k1
kobs
ˆ
‰OHÀŠ; K1 ˆ
ꢀ3†
k
k
À1
À1
The values of product K1k1 calculated from the experi-
mental results for derivatives 2b and 2c are K1k1 =
(6.30 Æ 0.04) Â 10À3 l molÀ1 sÀ1 and (3.18 Æ 0.03)
 10À3 l molÀ1
s
À1, respectively. The decrease in the
hydrolysis rate of 2c compared with compound 2b can
then be explained by both steric and inductive field
effects of the a-methyl group, which increase the electron
Figure 3. Dependence of logarithm of observed rate
constant on logarithm of sodium hydroxide concentration
for hydrolysis reaction 2b → 3b in aqueous medium 2*) and
in 10% 2v/v) DMSO 2*)
Scheme 4
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

´
M. SEDLAK ET AL.
172
The same character of the dependence of observed rate
constant on concentration of OHÀ ion was also observed
in the hydrolysis of 2d in aqueous sodium hydroxide,
where only the reaction path via dianion In2À is
significant and the values of the products are K1k2 =
(8.98 Æ 0.08) Â 10À3 l molÀ1 s1 and K1k3 = (1.49 Æ
0.02) Â 10À1 l2 molÀ2 s1. Hence it can be stated that
the substitution by bromine at the a-position of the g-
lactam ring of derivative 2d also caused an increase in
acidity of intermediate InÀ. A further increase in acidity
of intermediate InÀ was due to replacement of the nitrile
group by nitro in the aromatic moiety, as can be seen
from the dependence of logkobs vs logcNaOH for 2a. It was
found that the slope of this dependence decreased from 2
to 1. This change in the reaction order in reagent can be
interpreted by a change in the rate-limiting step: the OHÀ
ion-catalysed decomposition of tetrahedral intermediate
InÀ was so much accelerated with increasing concentra-
tion of OHÀ ion (cNaOH >0.1 mol lÀ1) that the rate-
limiting step shifted to the formation of this intermediate.
Consequently, the observed rate constant can be
expressed as follows:
Scheme 5
twice as fast [K1k1 = (1.11 Æ 0.06) Â 10À2 l molÀ1
s
À1].
If the basicity of the medium is increased by using a
solution of sodium hydroxide in 10% (v/v) aqueous
DMSO, then the reaction order in OHÀ ion is increased,
and hence also the slope of dependence of logkobs vs
logcNaOH for the hydrolysis of derivative 2b is increased
to 2 (see Fig. 3). This enhancement of the basicity of the
medium caused a relative increase in the acidity of
intermediate17 InÀ, which means that the second reaction
path became significant, i.e. the hydroxide ion-catalysed
decomposition of intermediate InÀ (the reaction pathway
via intermediate In2À) (see Schemes 3 and 5).
From the experimental values and with help of
multiple linear regression, it was possible to find that
the observed rate constant of hydrolysis of derivative 2b
in 10% (v/v) aqueous DMSO obeys the following rate
equation:
2
b‰OHÀŠ
2
kobs ˆ a‰OHÀŠ ‡ b‰OHÀŠ
ꢀ4†
kobs
ˆ
ꢀ5†
1 ‡ c‰OHÀŠ
From Schemes 3 and/or 5 it follows that the regression
parameters a and b correspond to the products K1k2 and
K1k3, respectively (the last constant being k3 = K3k'3).
The calculated values of the products K1k2 and K1k3 are
(2.19 Æ 0.05) Â 10À3 l molÀ1 s1 and (2.17 Æ 0.02) Â
10À2 l2 molÀ2 s1, respectively.
Applying multiple non-linear regression to the experi-
mental kobs values, we calculated the values of regression
coefficients b and c, which from their meaning corre-
spond to the values of K1k3 = (9.63 Æ 0.05) Â 102 l molÀ2
sÀ1 and k3/kÀ1 = (1.40 Æ 0.04) Â 102 l molÀ1, respec-
tively (Scheme 3). From the ratio of k3/kÀ1 it follows that
the OHÀ ion-catalysed decomposition of tetrahedral
intermediate InÀ is 140 times faster than its formation.
The introduction of bromine at the a-position results in
acceleration of all reaction steps of hydrolysis and in an
increase in acidity of intermediate InÀ (Scheme 3). This
in turn results in overall acceleration of hydrolysis of
bromo derivative 2a (by two orders of magnitude in
0.05 mol lÀ1 NaOH) compared with that of 2b.
Figure 4 presents the experimental points of the
dependence of the observed rate constant on concentra-
tion of sodium hydroxide for hydrolyses of 2a, 2b, and 2d
interlaced by curves corresponding to Eqns (3), (4) and
(5), respectively, the equations containing the calculated
regression parameters a, b and c.
Our suggested mechanism for the alkaline hydrolysis
of 1-phenylpyrrolidin-2-ones 2a–d agrees with the
former mechanism14,15 suggested for solvolyses of
tertiary amides and secondary lactams. The differences
in the kinetic courses of hydrolysis are determined by
differences in the structures of the individual transition
states of the rate-limiting step. Owing predominantly to
polar and steric effects, the transition state of hydrolysis
of bromo derivative 2a resembles the transition state of
hydrolysis of acetanilides.3 On the other hand, the
Figure 4. Dependence of observed rate constant 2kobs) on
sodium hydroxide concentration 2cNaOH) for hydrolyses in
aqueous medium: 2a → 3a 2*), 2b → 3b 2&) and 2d → 3d
2*)
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173

FORMATION AND DECOMPOSITION OF 1-PHENYLPYRROLIDIN-2-ONES
173
transition state of hydrolysis of methyl derivative 2c
resembles more the transition state of hydrolysis of N-
phenyl-b-lactams.3
by the Max-Planck Society and the Grant Agency of the
Czech Republic (Grant No. 203/01/0227).
REFERENCES
CONCLUSION
1. Heinze-Krauss I, Angehrn P, Guerry P, Hebeisen P, Hubschwerlen
C, Kompis I, Page MGP, Richter HGF, Runtz V, Stalder H, Weiss
U, Wei CC. J. Med. Chem. 1996; 39: 1868–1871.
2. Olson RE, Sielecki TM, Wityak J, Pinto DJ, Batt DG, Frietze WE,
Liu J, Tobin AE, Orwat MJ, Di Meo SV, Houghton GC, Lalka GK,
Mousa SA, Racanelli AL, Hausner EA, Kapil RP, Rabel HR,
Thoolen MJ, Reilly TM, Anderson PS, Wexler RR. J. Med. Chem.
1999; 42: 1178–1192.
3. Bowden K, Bromley K. J. Chem. Soc. Perkin Trans. 2 1990; 2103–
2109.
4. Heathcock CH, Kath JC, Ruggeri RB. J. Org. Chem. 1995; 60:
1120–1130.
5. Scrimin P, D’Angeli F, Veronese AC. Synthesis 1982; 586–587.
6. Bo¨hrer G, Bo¨hrer P, Knorr R. Chem. Ber. 1990; 2167–2172.
7. Stirling CJM. J. Chem. Soc. 1960; 255–262.
8. Cunningham BA, Schmir GF. J. Org. Chem. 1966; 31: 3751–3754.
9. Brown KL, Zahonyi-Budo E. J. Am. Chem. Soc. 1982; 104: 4117–
4124.
The ring closure of substituted 4-chloro-N-phenylbuta-
namides to 1-phenylpyrrolidin-2-ones is accelerated by
substituents at the a-position of the butanamide skeleton.
A methyl group at this position accelerates the cycliza-
tion more than bromine does, since the latter substituent
negatively affects the nucleophilicity of the intramolecu-
lar reagent. The ring closure reactions of a-bromo
derivatives 1a and 1d unambiguously prefer formation
of a g-lactam ring to formation of an a-lactam. With all
the derivatives, it is possible to separate the kinetics of
ring closure from those of consecutive hydrolysis in
aqueous medium. The greatest difference between the
rates of ring closure and consecutive hydrolysis was
found with methyl derivatives 1c and 2c. The solvolysis
of substituted 1-phenylpyrrolidin-2-ones follows the
same mechanism as the solvolysis of N-alkylamides.
The hydrolysis of bromo derivatives 2a and 2d is not
accompanied by splitting off of bromide anion: only the
pyrrolidine ring is opened to give the respective
substituted 4-aminobutanoic acid as in the other cases,
the bromo substituent at the a-position of the g-lactam
ring accelerating all reaction steps of hydrolysis.
10. Headley AD, Starnes SD, Wilson LY, Famini GR. J. Org. Chem.
1994; 59: 8040–8046.
11. Mandolini L. Adv. Phys. Org. Chem. 1986; 22: 1–111.
12. Abdallah JM, Moodie RB. J. Chem. Soc., Perkin Trans. 2 1983;
1243–1249.
ˇ
13. Steˇrba V, Panchartek J. Reaction Kinetics in Organic Chemistry.
University of Pardubice: Pardubice, 1995.
14. Talbot RJE. In Comprehensive Chemical Kinetics, Vol. 10,
Bamford CH, Tipper CFH (eds). Elsevier: Asterdam, 1972; chapt.
3.
15. Challis BC, Challis JA. In The Chemistry of Amides, Zabicky J
(ed). Interscience: London, 1970.
16. Moreau JM, Annez de Taboada M, Van Brandt P, Bruylants A.
Tetrahedron Lett. 1970; 1255–1258.
ˇ
17. Kava´lek J, Krampera F, Steˇrba V. Collect. Czech. Chem. Commun.
Acknowledgements
1976; 41: 1685–1691.
18. Blackburn GM, Plackett JD. J. Chem. Soc., Perkin Trans. 2 1972;
The authors acknowledge the financial support provided
1366–1371.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 165–173