Evidence of the structure of 4-(4′-acetoxybenzylidene)-2-methyl-5-oxazolone and its phenylpropenoic acid derivatives

Article abstract of DOI:10.1016/S0022-2860(02)00572-0

4-(4′-Acetoxybenzylidine)-2-methyl-5-oxazolone (Z-isomer) was synthesized, which upon basic hydrolysis yielded the unexpected 2-acetoxy-3-(p-hydroxyphenyl)-propenoic acid and treatment with acetic acid formed 2-acetoxy-3-(p-acetoxyphenyl)-propenoic acid. The structures of the three compounds were assigned by NMR spectroscopy. An X-ray crystallographic study of the starting azlactone confirmed its structure and supported the fact that the Z-configuration was retained during the synthesis of the phenylpropenoic acid derivatives.

Full text of DOI:10.1016/S0022-2860(02)00572-0

Journal of Molecular Structure 648 (2003) 61–67
www.elsevier.com/locate/molstruc
0
Evidence of the structure of 4-(4 -acetoxybenzylidene)-2-methyl-5-
oxazolone and its phenylpropenoic acid derivatives
a,
P.P. Haasbroek , D.W. Oliver , A.J.M. Carpy
b
c
*
a
Department of Pharmacy and Pharmacology, University of the Witwatersrand, 7 York Road, Parktown 2193, South Africa
b
Department of Pharmacology, Potchefstroom University for C.H.E., Potchefstroom 2520, South Africa
c
Laboratoire de Physico- and Toxico-Chimie des Syst e` mes Naturels (LPTC), UMR 5472 CNRS, Universit e´ de Bordeaux I,
51 Cours de la Lib e´ ration, 33405 Talence Cedex, France
3
Received 5 August 2002; revised 21 October 2002; accepted 21 October 2002
Abstract
0
4
-(4 -Acetoxybenzylidine)-2-methyl-5-oxazolone (Z-isomer) was synthesized, which upon basic hydrolysis yielded the
unexpected 2-acetoxy-3-( p-hydroxyphenyl)-propenoic acid and treatment with acetic acid formed 2-acetoxy-3-( p-
acetoxyphenyl)-propenoic acid. The structures of the three compounds were assigned by NMR spectroscopy. An X-ray
crystallographic study of the starting azlactone confirmed its structure and supported the fact that the Z-configuration was
retained during the synthesis of the phenylpropenoic acid derivatives.
q 2003 Elsevier Science B.V. All rights reserved.
Keywords: NMR; X-ray; Azlactones; Phenylpyruvic acids; Phenylpropenoic acids; Macrophage migration inhibitory factor
1
. Introduction
on their keto/enol ratio, with the enol form always
predominating [7]. NMR studies have further shown
that the solvent and temperature also influence
the keto/enol ratio [8]. In the case of the synthesis
of p-hydroxyphenylpyruvic acid, interesting results
This study is a continuation of our interest in the
synthesis of phenylpyruvic acids and derivatives via
the basic or acidic hydrolysis of their corresponding
azlactones. Compounds of this type are potential
substrates or inhibitors of the phenylpyruvate tauto-
merase activity catalyzed by the macrophage
migration inhibitory factor (MIF) [1–6]. Our study
on phenylpyruvic acids has so far revealed that
0
were observed. The basic hydrolysis of 4-(4 -acetox-
ybenzylidine)-2-methyl-5-oxazolone (Z-isomer) 2 did
not yield the expected phenylpyruvic acid, but its enol
acetate 3 (Synthetic scheme). Conversion of 2 to the
enol acetate derivative 4 (Synthetic scheme), by
treatment with acetic acid, a method developed in our
laboratory for the preparation of enol acetates [9],
made it possible to confirm the structure of the enol
acetate 3 by NMR spectroscopy. Compound 4 not
only has the enol acetate but also an aromatic acetate
group. This study, supported by previous studies
different substituents (e.g. Cl, NO …) as well as
2
their positions in the aromatic ring have an influence
*
Corresponding author. Tel.: þ27-11-717-2268; fax: þ27-11-
42-4355.
E-mail address: haasbroekpp@therapy.wits.ac.za (P.P.
Haasbroek).
6
0022-2860/03/$ - see front matter q 2003 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(02)00572-0

62
P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
Scheme 1.
[
9–11], has now shown that NMR spectroscopy can
110 8C. The reaction mixture was poured into a
mixture (120 ml) of equal volumes of ethanol and
water and stirred for 30 min. The product that
precipitated out was filtered off, washed with the
ethanol/water mixture and dried (15.2 g; 0.062 mol).
Recrystallization from methanol gave yellow crystals
of 2, m.p. 140–143 8C.
not only be used to distinguish between these two
kinds of esters but also to confirm the structures of
these types of phenylpyruvic acid (phenylpropenoic
acid) derivatives. An X-ray crystallographic study
was also performed on 2 in order to confirm the
spectroscopic findings (Scheme 1).
2.1.2. Synthesis of 2-acetoxy-3-( p-hydroxyphenyl)-
propenoic acid, 3
2
. Experimental
The azlactone 2 (2.0 g; 0.0082 mol) was boiled
gently under reflux in a sodium hydroxide solution
2
.1. Synthesis
(20%; 20 ml). After 2 h, the mixture was cooled and
The synthesis of the azlactone (Z ), 2, involved the
acidified with hydrochloric acid (6 M) to a pH of 1.4,
stirred well and allowed to cool in a refrigerator
overnight. The crystals that formed were filtered off,
washed with water and dried (0.73 g; 0.0033 mol).
Recrystallization from water gave off-white crystals
of 3, decomposing at 198–200 8C.
Erlenmeyer method [9,11–13] for the preparation of
aromatic azlactones, starting from p-hydroxybenzal-
dehyde and N-acetylglycine. Basic hydrolysis of 2,
surprisingly yielded 3, the enol acetate derivative
instead of the enol form of the phenylpyruvic acid,
which we expected to get. The diacetoxy derivative, 4,
was obtained from 2 by treatment with acetic acid.
2.1.3. Synthesis of 2-acetoxy-3-( p-acetoxyphenyl)-
propenoic acid, 4
0
2.1.1. Synthesis of 4-(4 -acetoxybenzylidene)-2-
methyl-5-oxazolone, 2
The azlactone 2 (3.0 g; 0.012 mol) was boiled
under reflux in acetic acid (90%, 30 ml) for 4 h. Water
(30 ml) was added and the mixture cooled in ice. The
crystals that formed were filtered off, washed with
water and dried (1.87 g; 0.0071 mol). Recrystalliza-
tion from a benzene/methanol (1:2) mixture gave
A mixture of p-hydroxybenzaldehyde (12.2 g;
.1 mol), N-acetylglycine (11.7 g; 0.1 mol), anhy-
0
drous sodium acetate (16.4 g; 0.2 mol) and acetic
anhydride (50 ml) was stirred for 2.5 h at 105–

P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
63
Table 1
1
H NMR data (d, ppm) of 2, 3 and 4
Hydrogen atom
2
3
4
a
H-2, H-6 (Ar)
8.21 (d, J ¼ 8:7)
7.51 (d, J ¼ 8:7)
7.67 (d, J ¼ 8:5)
0
0
0
H-3, H-5 (Ar)
7.27 (d, J ¼ 8:7)
6.81 (d, J ¼ 8:7)
7.17 (d, J ¼ 8:5)
0
0
0
Ar–CHyC
7.23 (s)
2.39 (s)
2.30 (s)
–
7.22 (s)
7.25 (s)
–
CH
CH
CH
3
3
3
–CyN
–
–CyO (Ar)
–CyO (En)
–
2.27 (s)
2.01 (s)
2.00 (s)
a
0
Ar: aromatic, En: enolic, d: doublet, s: singlet, J in Hz.
cream-coloured crystals of 4, m.p. 227–228 8C
decomp.)
structure with the atomic numbering scheme
adopted, is shown in Fig. 1.
(
2
.2. Spectroscopic analysis
3. Results
NMR analyses of compounds 2, 3 and 4 were
performed on a Bruker Avance 300 spectrometer,
using DMSO-d6 as solvent and TMS as internal
standard (Tables 1 and 2). The melting points
3
.1. Spectroscopic analysis
1
The H NMR data of 2, 3 and 4 appear in Table 1.
(uncorrected) were determined on an electrothermal
melting point apparatus.
The single proton signal at d ¼ 7.3 for 2 is the
characteristic absorption of the benzylic proton (Ar–
CHyC), confirming the Z-configuration of this
azlactone. The chemical shift value of the benzylic
proton can also be used to distinguish between the Z
and E isomers (further down field at d < 7.5) of these
2
.3. X-ray analysis
Yellow crystals of 2 were obtained from
methanol. The selected crystal (dimensions
3
0
.37 £ 0.22 £ 0.07 mm ) was mounted on an Enraf-
Nonius CAD-4 diffractometer using graphite-monochro-
mated CuKa radiation. Lattice parameters were
determined by least-squares refinement of 25
reflections with 188 , u , 328. The intensities
were collected with v 2 2u scan mode, up to
u ¼ 658. The intensities of two standard reflections,
measured every 908, showed no significant devi-
ation. The data were corrected for Lorentz and
polarization effects but not for absorption
Table 2
13
C NMR data (d, ppm) of 2, 3 and 4
Carbon atom
2
3
4
CHyC–CyO
167.2 (s)
166.7 (s)
168.8 (s)
–
–
–
CH
CH
CH
3
3
3
–CyN
–
–
a
–CyO (Ar)
–
169.0 (s)
169.3 (s)
166.3 (s)
127.3 (s)
130.9 (d)
122.0 (d)
150.8 (s)
131.3 (s)
130.1 (d)
20.8 (q)
22.5 (q)
–
–CyO (En)
169.40 (s)
166.7 (s)
124.0 (s)
131.9 (d)
115.5 (d)
158.7 (s)
124.6 (s)
132.6 (d)
–
COOH
–
2
1
(
m ¼ 8.81 cm ). The structure was solved by
C-1 (Ar)
132.3 (s)
133.1 (d)
122.3 (d)
152.2 (s)
130.6 (s)
128.7 (d)
20.7 (q)
–
C-2, C-6 (Ar)
C-3, C-5 (Ar)
C-4 (Ar)
direct methods using MolEN [14] and refined
anisotropically (Table 3). It was determined that
the structure corresponded to the Z azlactone. The
final residuals were R ¼ 0.0633 (all data) and
R ¼ 0.0486 [I . 2sðIÞ]. The final fractional coor-
dinates are listed in Table 4. The bond lengths and
angles are given in Tables 5 and 6. Selected torsion
angles are shown in Table 7. The molecular
Ar–CHyC
Ar–C HyC
CH
CH
CH
3
3
3
–CyO (Ar)
–CyO (En)
–CyN
22.6 (q)
–
15.2 (q)
a
Ar: aromatic, En: enolic, s: singlet, d: doublet, q: quartet.

64
P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
Table 3
Crystal data and structure refinement for 2
Table 5
˚
Bond lengths (A) for 2
Chemical formula
Formula weight
Temperature
C
13
H
11NO
4
C(1)–C(6)
1.393(3)
1.396(3)
1.458(3)
1.379(3)
1.380(3)
1.375(3)
1.404(2)
1.382(3)
1.340(3)
1.404(3)
1.474(3)
1.193(3)
1.387(3)
1.381(3)
1.279(3)
1.468(3)
1.351(3)
1.181(3)
1.496(3)
245.23
C(1)–C(2)
293(2) K
˚
1.54180 A
C(1)–C(7)
Wavelength
C(2)–C(3)
Crystal system
Space group
Orthorhombic
C(3)–C(4)
Pbca
C(4)–C(5)
˚
Unit cell dimensions
a ¼ 26.522(2) A, a ¼ 908
C(4)–O(15)
C(5)–C(6)
˚
b ¼ 11.783(1) A, b ¼ 908
˚
c ¼ 7.457(1) A, g ¼ 908
C(7)–C(8)
3
˚
Volume
2330.4(4) A
C(8)–N(12)
C(8)–C(9)
Z
8
2
3
Calculated density
Absorption coefficient
F(000)
1.398 mg m
C(9)–O(13)
C(9)–O(10)
O(10)–C(11)
C(11)–N(12)
C(11)–C(14)
O(15)–C(16)
C(16)–O(17)
C(16)–C(18)
2
1
0.881 mm
1024
3
Crystal size
Theta range
Index ranges
0.37 £ 0.22 £ 0.07 mm
3–658
2 # h # 30, 0 # k # 13, 0 # l # 8
Reflections collected/unique 1844/1844
Completeness to 2u ¼ 658 93.0%
Max. and min. transmission 0.9409 and 0.7365
2
Refinement method
Full-matrix least-squares on F
Data/restraints/parameters
2
Goodness-of-fit on F
1844/0/165
1.050
2
w
2
Final R indices [I . 2sðIÞ] R ¼ 0.0486, R
¼ 0.1266
¼ 0.1394
R indices (all data)
R ¼ 0.0633, R
w
Extinction coefficient
Largest diff. peak and hole
0.0059(6)
˚
0.299 and 20.280 eA
Table 6
23
Bond angles (8) for 2
C(6)–C(1)–C(2)
C(6)–C(1)–C(7)
C(2)–C(1)–C(7)
C(3)–C(2)–C(1)
C(2)–C(3)–C(4)
C(5)–C(4)–C(3)
C(5)–C(4)–O(15)
C(3)–C(4)–O(15)
C(4)–C(5)–C(6)
C(5)–C(6)–C(1)
C(8)–C(7)–C(1)
C(7)–C(8)–N(12)
C(7)–C(8)–C(9)
N(12)–C(8)–C(9)
O(13)–C(9)–O(10)
O(13)–C(9)–C(8)
O(10)–C(9)–C(8)
C(11)–O(10)–C(9)
N(12)–C(11)–O(10)
N(12)–C(11)–C(14)
O(10)–C(11)–C(14)
C(11)–N(12)–C(8)
C(16)–O(15)–C(4)
O(17)–C(16)–O(15)
O(17)–C(16)–C(18)
O(15)–C(16)–C(18)
118.2(2)
118.4(2)
123.4(2)
121.3(2)
118.7(2)
121.7(2)
114.7(2)
123.3(2)
119.0(2)
121.0(2)
128.4(2)
129.3(2)
123.2(2)
107.3(2)
121.9(2)
133.2(2)
104.9(2)
105.9(2)
115.4(2)
128.8(2)
115.7(2)
106.4(2)
121.7(2)
123.7(2)
125.9(2)
110.4(2)
Table 4
4
Atomic coordinates ( £ 10 ) and equivalent isotropic displacement
2
3
˚
parameters (A £ 10 ) for 2. U(eq) is defined as one third of the
trace of the orthogonalized Uij tensor
x
y
z
U(eq)
C(1)
7935(1)
7759(1)
7259(1)
6929(1)
7090(1)
7594(1)
8453(1)
8797(1)
9299(1)
9512(1)
9162(1)
8753(1)
9521(1)
9309(1)
6419(1)
6136(1)
6290(1)
5609(1)
3849(2)
2867(2)
2750(2)
3619(2)
4611(2)
4728(2)
3991(2)
3188(2)
3437(2)
2382(1)
1593(2)
2003(2)
4298(2)
399(2)
21111(3)
2273(3)
205(3)
38(1)
41(1)
41(1)
40(1)
44(1)
44(1)
43(1)
41(1)
52(1)
57(1)
48(1)
45(1)
79(1)
67(1)
52(1)
50(1)
91(1)
62(1)
C(2)
C(3)
C(4)
2195(3)
2980(3)
21414(3)
21723(3)
22068(3)
22822(4)
23155(3)
22588(3)
21951(3)
23118(3)
22791(5)
304(2)
C(5)
C(6)
C(7)
C(8)
C(9)
O(10)
C(11)
N(12)
O(13)
C(14)
O(15)
C(16)
O(17)
C(18)
3612(1)
2660(2)
1783(2)
2893(3)
203(4)
2337(4)
815(4)

P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
65
Table 7
the phenylpyruvic acid. The benzylic carbon of the
enol acetate absorbs at d ¼ 132.6 (chemical shift
values of d ¼ 120–133 were reported for similar enol
acetates) while the same proton of phenylpyruvic
acids (enol form) absorbs higher upfield (104–
Selected torsion angles (8) for 2
C(6)–C(1)–C(7)–C(8)
C(1)–C(7)–C(8)–C(9)
C(5)–C(4)–O(15)–C(16)
C(4)–O(15)–C(16)–C(18)
2 158.3(2)
174.9(2)
145.2(2)
180.0(2)
112 ppm) [8]. These values are also in agreement
with values obtained for the enol acetate of the 4-
hydroxy-3-methoxy derivative [10].
The structure of the diacetoxy derivative, 4, as
kind of 2-methyl azlactones [15,16]. The absorption
of this proton can be easily observed in the methyl
azlactones than in the phenyl azlactones, as the
absorption of the extra aromatic protons in the latter
compounds tend to mask the absorption of the
benzylic proton.
1
13
enol acetate, was confirmed by its H and
C
NMR spectra, which further supports the structure
1
of 3. H NMR spectral values of the methyl
protons of the aromatic acetate group (d < 2.30)
and the enolic acetate group (d < 2.00) can be used
to confirm the presence of, or distinguish between
With compound 3, where we expected the enol
1
form of 4-hydroxyphenylpyruvic acid to form, its H
1
3
NMR spectrum clearly showed that, instead, the enol
acetate was formed. The three-proton singlet at
d ¼ 2.00 is characteristic of the methyl protons of
the enol acetate. The benzylic proton of the enol
acetate absorbs at d ¼ 7.22 which is similar to values
found for similar derivatives (d ¼ 7.1–7.6) [9–11].
The benzylic proton of phenylpyruvic acids (enol
form) absorbs higher upfield (d ¼ 6.37–6.8). This is
these two groups. The C NMR chemical shift
values of these methyl carbons can also be
effectively used to distinguish between these two
esters, as the aromatic ester gives a quartet at
d < 20.7 and the enolic ester, a quartet at d < 22.6.
The chemical shift values of the carbonyl carbons
of these esters are very close and could not be used
to distinguish between them.
1
3
further confirmed by the C NMR spectrum of 3
which shows the absorption of 11 carbon atoms and
not nine as it would have been the case for
The similar absorption values for the benzylic
proton at d ¼ 7.23, 7.22 and 7.25 for 2, 3 and 4,
respectively, can therefore be used to confirm the
Fig. 1. ORTEP view of 2.

66
P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
existence of the Z-configuration of such compounds.
the existence of the Z isomer and that the ester group
is in the same plane as the aromatic ring.
1
3
The
benzylic carbon at d ¼ 128.7, 132.6 and 130.1 for
, 3 and 4, respectively, also seem to be good
C chemical shift values for the same
Finally, the macrophage MIF, a cytokine impli-
cated in a number of immune and inflammatory
processes, is known for interconverting the enol and
keto isomers of both phenylpyruvate and p-hydro-
xyphenylpyruvate. Crystal structures of MIF com-
plexed with competitive inhibitors have shown that
the substrates exist as enols (E-isomers) [17,18].
Following a molecular modeling process, work is in
progress to synthesize phenylpyruvic acid derivatives
and parent compounds with the (E )-configuration as
potential inhibitors of the phenylpyruvic tautomerase
activity of MIF.
2
indicators for the confirmation of the Z-configur-
ations.
3
.2. X-ray analysis
Bond lengths and angles of the azlactone 2 have the
suspected values. The Z-configuration was confirmed
by this single crystal X-ray analysis. The N(12)
nitrogen atom and the aromatic ring are on the same
side of the C(7)–C(8) double bond. The relative
orientation of the ester group versus the aromatic ring
is defined by the two torsion angles C(5)–C(4)–
O(15)–C(16) ¼ 145.2(2)8 and C(4)–O(15)–C(16)–
C(18) ¼ 180.0(2)8.
Acknowledgements
The authors are grateful to Prof. J.M. L e´ ger
Laboratoire de Chimie Analytique, UFR des
(
Sciences Pharmaceutiques, Universit e´ de Bordeaux
II) for collecting the X-ray crystal data of 2.
4
. Discussion
The basic hydrolysis of the azlactone (Z ) 2, yields
the enol acetate derivative 3. The same phenomenon
is observed with the 4-hydroxy-3-methoxy derivative
[
9]. The enolic hydroxyl group seems to react readily
References
with the acetate group that becomes available during
hydrolysis of the aromatic acetate group and the
azlactone ring. Whereas the aromatic acetate is easily
hydrolyzed, the enol acetate must be quite stable to
form under these conditions and to resist hydrolysis.
However, we found that the 3-hydroxy-4-methoxy
substituted derivative does form the enol form of the
phenylpyruvic acid when hydrolyzed under similar
conditions [8]. It appears that the enol acetate is less
stable when the phenolic hydroxyl is in the 3-position
and therefore hydrolyzes under these conditions. This
was confirmed using an AMPAC energy calculation
on both enol acetates, of 4-hydroxy and 3-hydroxy
compounds, the former appearing to be the most
stable one.
[1] T. Calandra, R. Bucula, Crit. Rev. Immunol. 17 (1997) 77.
[
2] S.L. Stamps, A.B. Taylor, S.C. Wang, M.L. Hackert, C.P.
Whitman, Biochemistry 39 (2000) 9671.
[
3] M. Orita, S. Yamamoto, N. Katayama, M. Aoki, K. Takayama,
Y. Yamagiwa, N. Seki, H. Suzuki, H. Kurihara, H. Sakashita,
M. Takeuchi, S. Fujita, T. Yamada, A. Tanaka, J. Med. Chem.
44 (2001) 540.
[
4] A. Dios, R.A. Mitchell, B. Aljabari, J. Lubetsky, K. O’Connor,
H. Liao, P.D. Senter, K.R. Manogue, E. Lolis, C. Metz, R.
Bucala, D.J.E. Callaway, Y. Al-Abed, J. Med. Chem. 45
(
2002) 2410.
[
[
5] P.D.Senter,Y.Al-Abed,C.N.Metz,F.Benigni,R.A.Mitchell,J.
Chesney, J. Han, C.G. Gartner, S.D. Nelson, G.J. Todaro, R.
Bucala, Proc. Natl. Acad. Sci. USA 99 (2002) 144.
6] J.B. Lubetsky, A. Dios, J. Han, B. Aljabari, B. Ruzsicska, R.
Mitchell, E. Lolis, Y. Al-Abed, J. Biol. Chem. 277 (2002)
2
4976.
The structure of 4 was clearly established using
NMR spectroscopy. However, the mechanism for its
formation remains unclear. Further investigation is
currently in progress to establish the origin of the enol
acetate group, i.e. whether its source is from the
acetic acid or from the azlactone ring. The data from
the X-ray crystal study of the azlactone 2 confirmed
[
[
[
7] A.J.M. Carpy, P.P. Haasbroek, J. Ouhabi, D.W. Oliver, J. Mol.
Struct. 520 (2000) 191.
8] P.P. Haasbroek, PhD Thesis, Potchefstroom University for
C.H.E., Potchefstroom, South Africa (1997).
9] P.P. Haasbroek, D.W. Oliver, A.J.M. Carpy, J. Chem. Crystal-
logr. 28 (1998) 139.
[
10] P.P. Haasbroek, D.W. Oliver, A.J.M. Carpy, J. Chem. Crystal-
logr. 28 (1998) 193.

P.P. Haasbroek et al. / Journal of Molecular Structure 648 (2003) 61–67
67
[
11] D.W. Oliver, P.P. Haasbroek, J.M. L e´ ger, A.J.M. Carpy,
[15] R. Glaser, M. Twaik, Tetrahedron Lett. 15 (1976) 1219.
J. Chem. Crystallogr. 24 (1994) 665.
[16] A.P. Morgenstern, C. Schutij, W.T. Nauta, Chem. Commun.
(1969) 321.
[
[
12] H.E. Carter, Org. React. 3 (1946) 199.
13] P.P. Haasbroek, D.W. Oliver, A.J. Carpy, J. Chem. Crystal-
logr. 28 (1998) 811.
[17] J.B. Lubetsky, M. Swope, C. Dealwis, P. Blake, E. Lolis,
Biochemistry 38 (1999) 7346.
[
14] MolEN, An Interactive Structure Solution Procedure, Enraf-
[18] A.B. Taylor, W.H. Johnson Jr., R.M. Czerwinski, H.S. Li,
M.L. Hackert, C.P. Whitman, Biochemistry 38 (1999) 7444.
Nonius, Delft, The Netherlands, 1990.