Assisted hydrolysis of cis-2-(3-phenylthioureido)cyclopentane-carbonitrile in alkaline solution. Solvent dependent switch from hydrolysis to rearrangement of the iminothiooxopyrimidine intermediate

Article abstract of DOI:10.1139/v99-216

The cis and trans isomers of 2-(3-phenylthioureido)cyclopentanecarbonitrile, 1, and the respective carboxamides, 3, and acids, 4, have been prepared. Acid cyclization of both nitriles, faster with the cis isomer, gave the more stable cis-2-thiooxo-cyclopenta[d]pyrimidin-4-one, 7. In base cis-1 formed the cis 4-imino-2-thiooxopyrimidine 2 which in aqueous alkali broke down via 3 to the acid 4; while in the presence of 66% acetonitrile 2 rearranged to the 4-phenyliminopyrimidine 5 to give as final product the thioureido acid 6 carrying no phenyl group. The ¹H NMR data for imino and phenylimino derivatives 2 and 5 showed strong bias for conformation A with 1-N pseudoaxial in the cyclopentane ring. Spectra of the E and Z isomers of the iminopyrimidine 2 under slow exchange could be recorded in DMSO-d₆. The phenylimino tautomer of 5 is observed in CD₃OD and in CDCl₃ with the E and Z isomers in a 1:1 ratio. In DMSO-d₆ the phenylamino tautomer 5a is also detected. The first process in aqueous KOH, the conversion of nitrile cis-1 into the imino intermediate 2, reaches an equilibrium which shifts towards the nitrile at higher alkalinities because of ionization of the phenylthioureido group (K_e = [2]/[1] = 2.43 and _pK_AH = 12.74). The cyclization of 1 to 2 is first order in [OH^-] while the slower breakdown of 2 is pH independent. The latter is 10⁴ times faster than the hydrolysis of acetonitrile evidencing substantial anchimeric assistance. The change in the reaction route towards the rearranged phenyliminopyrimidine 5 upon addition of acetonitrile can be caused by the lower dielectric constant favouring the elimination step leading to the intermediate isothiocyanate, and by increased activity of OH^- accelerating the (presumably) second order elimination step as opposed to the ph-independent hydrolysis of the imino derivative 2.

Full text of DOI:10.1139/v99-216

84
Assisted hydrolysis of cis-2-(3-
phenylthioureido)cyclopentane-carbonitrile in
alkaline solution. Solvent dependent switch from
hydrolysis to rearrangement of the
iminothiooxopyrimidine intermediate
Ergun Atay, Iva B. Blagoeva, Francis L. Chubb, John T. Edward,
Ivan G. Pojarlieff, and Maria M. Toteva
Abstract: The cis and trans isomers of 2-(3-phenylthioureido)cyclopentanecarbonitrile, 1, and the respective
carboxamides, 3, and acids, 4, have been prepared. Acid cyclization of both nitriles, faster with the cis isomer, gave the
more stable cis-2-thiooxo-cyclopenta[d]pyrimidin-4-one, 7. In base cis-1 formed the cis 4-imino-2-thiooxopyrimidine 2
which in aqueous alkali broke down via 3 to the acid 4; while in the presence of 66% acetonitrile 2 rearranged to the
1
4-phenyliminopyrimidine 5 to give as final product the thioureido acid 6 carrying no phenyl group. The H NMR data
for imino and phenylimino derivatives 2 and 5 showed strong bias for conformation A with 1-N pseudoaxial in the
cyclopentane ring. Spectra of the E and Z isomers of the iminopyrimidine 2 under slow exchange could be recorded in
DMSO-d6. The phenylimino tautomer of 5 is observed in CD3OD and in CDCl3 with the E and Z isomers in a 1:1
ratio. In DMSO-d6 the phenylamino tautomer 5a is also detected. The first process in aqueous KOH, the conversion of
nitrile cis-1 into the imino intermediate 2, reaches an equilibrium which shifts towards the nitrile at higher alkalinities
because of ionization of the phenylthioureido group (Ke = [2]/[1] = 2.43 and pKAH = 12.74). The cyclization of 1 to 2
is first order in [OH–] while the slower breakdown of 2 is pH independent. The latter is 104 times faster than the
hydrolysis of acetonitrile evidencing substantial anchimeric assistance. The change in the reaction route towards the
rearranged phenyliminopyrimidine 5 upon addition of acetonitrile can be caused by the lower dielectric constant
favouring the elimination step leading to the intermediate isothiocyanate, and by increased activity of OH– accelerating
the (presumably) second order elimination step as opposed to the pH-independent hydrolysis of the imino derivative 2.
Key words: anchimeric assistance, phenylthioureido nitriles, iminothiooxopyrimidine tautomers, alkaline hydrolysis,
solvent effects. 94
Résumé : On a préparé les isomères cis et trans du 2-(3-phénylthiouréido)cyclopentanecarbonitrile (1) et des
carboxamides (3) et acides (4) correspondants. La cyclisation des deux nitriles, plus rapide pour l’isomère cis, conduit
dans chaque cas à la cis-2-thiooxocyclopenta[d]pyrimidin-4-one (7), l’isomère le plus stable. En milieu basique,
l’isomère 1-cis conduit à la formation de la cis-4-imino-2-thiooxopyrimidine (2) qui, dans ces conditions, se décompose
en acide 4 par le biais de 3; toutefois, dans un milieu contenant 66% d’acétonitrile, le produit 2 se réarrange en
4-phényliminopyrimidine (5) qui se transforme pour fournir comme produit final l’acide thiouréido 6 où la groupe
1
phenyle est absent. Les données de RMN du H pour les dérivés imino et phénylimino 2 et 5 montrent que la confor-
mation A est fortement favorisée, avec le 1-N en position axiale par rapport au cyclopentane. En opérant dans le
DMSO-d6, on a pu enregistrer les spectres des isomères E et Z de l’iminopyrimidine dans des conditions d’échange
lent. Dans le CD3OD et dans le CDCl3, on peut observer le tautomère phénylamino de 5; les isomères E et Z se
trouvent dans un rapport de 1:1. On a aussi détecté le tautomère phénylamino 5a dans le DMSO-d6. Le premier
processus réactionnel observé dans le KOH aqueux, la conversion du nitrile 1-cis en intermédiaire imino 2 atteint un
équilibre qui, en raison de l’ionisation du groupe phénylthiouréido (Ke = [2]/[1] = 2,43 et pKAH = 12,74), se déplace
vers le nitrile à des niveaux d’alcalinité élevés. La cyclisation de 1 et 2 est du premier ordre en [OH–] alors que la
décomposition plus lente de 2 est indépendante du pH. Cette dernière réaction est 104 fois plus rapide que l’hydrolyse
de l’acétonitrile; ceci confirme l’importance de l’assistance anchimérique. Le changement de la voie réactionnelle
Accepted October 25, 1999.
E. Atay, I.B. Blagoeva, I.G. Pojarlieff,1 and M.M. Toteva. Institute of Organic Chemistry, Bulgarian Academy of Sciences, ul.
Acad. G. Bonchev block 9, Sofia 1113, Bulgaria.
F.L. Chubb and J.T. Edward. Department of Chemistry, McGill University, 801 Sherbrooke St., Montreal, QC H3A 2K6, Canada.
1Author to whom correspondence may be addressed. Telephone: +359-2-7334-116. Fax: +359-2-700-225.
e-mail: ipojarli@orgchm.bas.bg
Can. J. Chem. 78: 84–94 (2000)
© 2000 NRC Canada

Atay et al.
85
conduisant à plus de phényliminopyrimidine réarrangée 5 par addition d’acétonitrile peut être provoqué par une baisse
de constante diélectrique favorisant l’étape d’élimination qui conduit à l’intermédiaire isothiocyanate et par une activité
accrue des ions OH– qui augmente la vitesse de l’étape d’élimination que l’on croit être du deuxième ordre par opposi-
tion à l’hydrolyse du dérivé imino 2, qui est indépendante du pH.
Mots clés : assistance anchimérique, nitriles phénylthiouréido, tautomères des iminothiooxopyrimides, hydrolyse
alcaline, effets de solvant.
[Traduit par la Rédaction]
Scheme 1.
Atay et al.
Introduction
but the phenylthioureido derivatives 1 are readily separated
by column chromatography.
Ureido or thioureido nitriles, readily available from amino
nitriles, have been widely used (1) in the synthesis of
aminopyrimidines, aminoimidazoles, or thiazoles. On a pre-
parative level the ring-closure was found to be base cata-
lyzed (2). The products of attack of the ω-nitrogen of the
ureido group are formed from thioureido nitriles in neutral
and basic media (3).
We now report that the hydrolysis of a nitrile group is
anchimerically assisted by a suitably orientated β-phenyl-
ureido group as in cis-1. The latter compound, in 0.01 M
aqueous alkali, converts into the amide 3 104 times faster
than acetonitrile (4). The hydrolysis is facilitated by the for-
mation of the iminopyrimidine 2 as shown in Scheme 1.
However, hydrolysis of aminopyrimidine 2 to 3 and fur-
ther to the phenylthiouredo acid 4 was observed in aqueous
alkali with a low (1–2%) content of acetonitrile. When the
concentration of MeCN was increased to 66% (v/v) the ma-
jor products were the rearranged pyrimidine 5, hydrolyzing
ultimately to the thioureido acid 6 without the phenyl group.
The configurations of the thioureido nitriles 1 were as-
signed from their behaviour in 1:2 HCl. Upon refluxing,
the two isomers both cyclized in high yield to produce the
single product cis 2-thiodihydrouracil 7 (Scheme 2). The
latter has recently been obtained by cyclization of the re-
spective cis thioureido ethyl ester in hydrochloric acid solu-
tion (5) and assigned as cis from bandwidths of the signals
1
for 4a-H (ca. 25 Hz) and 7a-H (ca. 12 Hz) in the H NMR
(6). We proved this independently by double resonance ex-
periments2 which showed J4a,7a to be 6.5 Hz, which fell
within the usual range for cis coupling in dihydrouracils
(7). The 7a-H signal splittings due to the neighbouring
methylene hydrogens were considerably smaller than those
for 4a-H, indicating that the NH group is pseudoaxial in the
presumably half-chair conformation of the cyclopentane
ring. Cyclization of cis and trans 1 under milder conditions
(2 h in 1 M HCl at 50°C) gave 86% and 15% cis-7, respec-
tively. Whatever the reaction path from cis- or trans-1 to
cis-7, it can be safely assumed that the more readily
cyclizing isomer has the same configuration as the ring
product. Isomerization most likely occurs by acid catalyzed
enolization which requires prior protonation of the car-
bonyl or imino function. It has been previously shown (8)
that upon cyclization of ureido acids to dihydrouracils,
isomerization occurs with the more basic ring compounds.
Under the conditions used for obtaining the amides 3 on a
Results and discussion
Preparation and configuration of 2-phenylthio-
ureidocyclopentanecarboxylic acid and 4-oxo-2-thio-
oxooctahydrocyclopenta[d]pyrimidine derivatives
2-Aminocyclopentanecarbonitrile is obtained as a mixture
of cis and trans isomers which have not yet been separated,
2 Irradiation of H-6 of thiodihydrouracil decreased the 4a-H band by 6.5 Hz. Irradiation of 1-H sharpened the signal for 7a-H to a multiplet of
8 lines with couplings 6.6, 5.3, and 4.2 Hz. In the multiplet for 4a-H these couplings are 6.6, 7.8, and 9.0 Hz.
© 2000 NRC Canada

86
Can. J. Chem. Vol. 78, 2000
Scheme 2.
preparative scale (hydrolysis of the nitriles 1 in 60% sulfu-
ric acid) the cis nitrile gave cis amide, shown by TLC to be
free of the trans amide, while the trans nitrile hydrolyzed
to a mixture of cis and trans isomers. Since the trans nitrile
1 and amide 3 are certainly more stable than cis-1 and 3,
the isomerization must occur after the formation of a
bicyclic derivative more stable in the cis configuration.
Greater stability of the cis isomer of dihydrouracil 7 can
be understood by reference to the hydrindane system
(bicyclo[4,3,0]nonane) (9). The trans hydrindane is more
stable (∆H) by 1.1 kcal mol–1 as the result of two conflicting
forces: the unavoidable axial interactions in the cis isomer
and the mismatch of torsion angles between the five and six
membered rings. In a trans junction these can be accommo-
dated by unfavourable greater puckering. Because dihydro-
uracils are considerably planar (10) with the torsion angles
between the trans diequatorial bonds ca. 90°, trans fusion to
a cyclopentane with an endocyclic torsion angle of ca. 45°
would require considerable strain, which is apparently the
main factor in determining the relative stability of the two
isomers. Exclusive preference for cis-annelation has also
been observed in the formation of 3-aryl-2,4-dioxo-octahy-
drocyclopenta[d]pyrimidines in a ring expansion reaction (6).
Preparation of 4-imino and 4-phenylimino-2-thio-
oxooctahydrocyclopenta[d]pyrimidine
When the hydrolysis of cis thioureido nitrile 1 was moni-
tored by UV in aqueous alkaline media containing only
1–2% acetonitrile, a decrease of absorption at 245 nm due to
NHCSNHPh and an accompanying increase in absorption at
267 nm were observed. The latter reached a maximum after
which the absorption at 245 nm was restored in a slower
process. Detailed product analysis (vide infra) showed that
the intermediate formed broke down to amide cis-3 and this
in turn to cis-acid 4 with both compounds having the same
UV absorption as the initial nitrile. The intermediate, ex-
pected to be the iminopyrimidine 2, however, proved elusive.
The usual work up gave the rearranged product, the phenyl-
iminopyrimidine 5. Its structure was deduced from the prod-
ucts of hydrolysis: in base the thioureido acid 6, while in
acid the known thiodihydrouracil 8 (11) without the phenyl
group. Compounds 6 and 8 could be readily interconverted
in acid or base as appropriate. Under the conditions de-
scribed in the experimental section no isomerization was ob-
1
served because H NMR evidence demonstrated retention of
cis configuration.
Under conditions where imines similar to 2 have been
obtained from ureido nitriles our starting nitrile cis-1 either
remained unchanged (refluxing in benzene (3a,b)) or yielded
the rearranged phenylimino product 5 directly (refluxing in
methanol (3a) or in EtOH–EtONa (3c)). As the behaviour of
cis-1 in aqueous alkali could only be explained by the partic-
ipation of imine 2 as depicted in Scheme 1, an attempt was
Acid cis-4 was readily available by means of alkaline hy-
drolysis of cis dihydrouracil 7 while trans acid 4 was ob-
tained under milder acid hydrolysis of trans nitrile 1 along
with some cis dihydrouracil 7. Acid trans-4 has been iso-
lated in high yield from the ethyl ester in an attempt to
cyclize the latter under reflux in hydrochloric acid (5).
© 2000 NRC Canada

Atay et al.
87
Table 1. 13C NMR chemical shifts (in ppm) from TMS, in DMSO-d6 solution
(a) Derivatives of cyclopentanecarboxylic acid
Compound
COOH
CS
1-C
2-C
3-C
4-C
5-C
i-C
o-C
m-C
p-C
cis-4
cis-6
175.1
174.8
179.9
182.8
45.8
46.0
56.6
57.1
31.0
31.1
21.4
21.4
27.4
27.1
139.2
—
124.0
—
128.5
—
122.7
—
(b) Derivatives of 2-thio-1,2,3,4a,5,6,7,7a-octahydrocyclopenta[d]pyrimidines
4-C
2-C
4a-C
7a-C
5-C
6-C
7-C
i-C
o-C
m-C
p-C
cis-7
cis-5a
cis-8
169.2
153.4
169.3
179.6
176.6
177.2
43.0
36.6
41.6
54.5
55.5
55.4
27.9
29.7
27.7
21.5
21.5
21.6
32.0
31.4
32.2
139.3
148.3
—
128.2
120.4
—
129.5
128.8
—
127.4
122.7
—
aResonances of the major isomer.
made to obtain it in this medium. To avoid the inconvenience
of the limited solubility of cis-1 in water, the experiment
was carried out with 0.01 M KOH in a water–acetonitrile
1:2 mixture. Unexpectedly the reaction proceeded by the
completely different course shown in Scheme 2. This meant
that the rate of isomerization became faster than the rate of
hydrolysis of the imine 2 to amide 3. The procedure for ob-
taining the non-rearranged imine cis-2 described in the ex-
perimental was based on the kinetic study of the hydrolysis
of the nitrile 1. Conditions were chosen where formation of
cis-2 was rapid and, at the moment of its maximum forma-
tion, the mixture was quenched in acid. The reaction is re-
versible and in acid unconverted nitrile was extracted from
the aqueous solution. The water layer could be used as a
stock solution of cis-2. The latter was isolated as a solid by
rapid alkalization of the acid solution from which the nitrile
has been removed, extraction with methylene chloride, and
showed only one, apparently time averaged, set of signals.
In DMSO, however, spectra under fast and slow exchange
were observed with various samples. Under fast exchange 1-
H resonated at 9.26 ppm while =NH was hidden under the
multiplet for the m-H and p-H at ca. 7.35 ppm, integrating
for 4 protons. Under slow exchange 1-H gave rise to two
signals: 9.38 (0.4H) and 9.15 (0.6H) ppm. The signals for
=NH were at 8.59 (0.6H) and 6.58 (0.4H) ppm respectively.
The 2 ppm upfield shift can be attributed to the ring current
effect of the 3-phenyl group in the Z-isomer because the
benzene ring is not in the plane of the heterocyclic ring in
similar systems (14). Other samples showed intermediate ex-
change. On standing all of them converted cleanly into the
phenyliminopyrimidine 5, indicating that the initial spectra
dealt with the same compound. In view of the rapid conver-
sion in DMSO of 2 into 5 and the difficulty in controlling
chemical exchange, further investigation was discontinued.
1
evaporation in vacuo. The H NMR of the solid showed
With compound 5 two tautomers are possible: the phenyl-
imino and phenylamino isomers shown. Similar cyclic
acylated phenylamidines such as 3-aryliminoisoindolinones
have been found to be only in the phenylimino tautomeric
form (15). This is also the predominant form with phenyl-
iminopyrimidine 5. An IR band at 1664 cm–1 is observed in
CHCl3 which is within the characteristic range for exocyclic
C–N double bonds non-conjugated to an acyl group (16).
Another feature is the upfield chemical shifts of the o- and
p-protons and carbons of the phenyl ring which is due to
better conjugation with the electron pair on nitrogen than is
the case when interaction with an acyl group (C=S) is possi-
ble, as in 5a (15) (see Tables 1 and 2). However, in DMSO a
third set of signals could be detected which was assigned to
the amino form 5a. The assignment was supported by vari-
able temperature measurements. Notably a doublet of the 5a
set at 7.89 ppm did not exchange with D2O and coalesced
with the ortho protons upon heating. That this belongs to the
interchanging system is supported by the fact that at 383 K
the m- and p-protons had already coalesced to sharp triplets
while the signal for the o-protons at ca. 6.8 ppm was still
broad due to the greater chemical shift difference (>1 ppm
between the imino and amino forms). The greater signifi-
cance of the amino form compared to the above mentioned
isoindolinones can be attributed to the greater acidity of
thioamides compared to ordinary amides. The isomer ratio is
E-5:Z-5:5a = 7.5:1.5:1.
small admixtures of cis-1 and the rearranged product cis-5,
both most likely formed during the work up. In solution the
imino compound converted into the phenylimino one with a
rate dependant on the solvent. Rough estimates of the half
lives (in min) were: 16 in DMSO, 130 in CD3OD, and 400
in CDCl3 (monitored by NMR at room temperature).
The structure of cis-2 followed from the chemical trans-
formations discussed above. The other possible structure,
that of a iminothiazinone resulting from sulfur attack on the
nitrile group can be refuted from the 13C spectra of the final
product of the transformations in Scheme 2. As shown in Ta-
ble 1 the thioureido acid cis-6 shows resonances for CS and
COOH at 182.8 and 174.8 ppm respectively. It can be
readily shown that the thiazine would ultimately afford an
ureido thioacid for which resonances are expected at ca. 160
and 195 ppm for CO and CSOH (12). In DMSO the structur-
ally similar R*,R*-2-methyl-3-ureidobutanoic acid showed
resonances at 176.2 (COOH) and 158.5 (NHCONH2).3
Tautomers and configurations of the imines.
The amidine moiety in iminopyrimidine cis-2 is fixed and
allows no tautomers. However, two geometrical isomers
around the C=N bond are possible. According to Perrin (13)
it is especially unusual to see separate forms such as Z-2 and
E-2 (Scheme 2), since proton exchange interconverts them.
1
Accordingly in CD3OD and CDCl3 the H NMR spectrum
3 Unpublished.
© 2000 NRC Canada

88
Can. J. Chem. Vol. 78, 2000
1
Table 2. H NMR data for compounds in DMSO-d6, chemical shifts from TMS (ppm), band widths, and couplings (Hz).a
(a) Derivatives of 2-phenylthioureidocyclopentanecarboxylic acid
Compound
1-H
2-H
o-Hc
m-Hc
p-Hd
1′-NH
3′-NH
nitrile cis-1
in CDCl3
3.53 quart bw 20.5
3.62 bw 18.2
4.74 mb
7.50 d
e
7.32 t
7.47 t
7.303 t
7.303 t
7.297 t
7.305 t
7.297 t
—
7.11 t
e
8.06 d J ca. 7
6.19 d J 7.5
7.99 d J 7.3
7.64 d J 7.9
7.80b
9.60 s
4.902 bw 33.1
4.871 quin bw 29.7
4.804 mb
7.87 s
nitrile trans-1
amide cis-3 f
amide trans-3g
acid cis-4h
acid trans-4i
cis-6 j
3.060 quart bw 22.9
2.879 quart bw 22.2?
2.614 quart bw 22.8
2.993 quart bw 21.2
2.709 quart bw 23
2.924 quart bw 21.3
7.397 d
7.432 d
7.452 d
7.469 d
7.423 d
—
7.117 t
7.087 t
7.075 t
7.09 t
7.076 t
—
9.59 s
9.80 s
4.66b
9.41 s
4.891b
4.748b
4.71b
m
7.625 d J 8.3
7.93 d J ca. 6
7.47 d J 7
9.696 s
9.44 s
7.05 s 2H
(b) Derivatives of 2-thioxo-1,2,3,4a,5,6,7,7a-octahydrocyclopenta[d]pyrimidines
Compound
4a-H
7a-H
o-Hc
m-Hc
p-Hd
1-NH
3-NH
n
n
4-oxok cis-7
3.045 quart bw 23.3l
2.770 quart bw 23.2
3.039 sext bw 22.2
3.100 sext bw 24
3.118 sext bw 22.5
2.584 bw 21
4.052 quart bw 15.9m
3.833 quart bw 19.0
3.945b hw 11.5
4.042 hw 12.0
4.042 hw 12
3.799b hw 10
3.85
7.1 d
9.84
4-oxo cis-8
9.49
9.26 so
10.93
p
4-iminok cis-2
in CD3OD
7.088 d
7.18 d
7.37 t
7.317 t
7.456 t
q
7.35 t
q
in CDCl3
7.295 d
6.752 d 7.7
6.83 d
7.21 sr
8.87 s
8.9 7s
7.77 s
r
4-phenylimino cis-5-Es
4-phenylimino cis-5-Zs
5as
7.299 t
7.029 t
10.51s
9.99s
9.43 s
2.74
7.378
t
7.109
t
2.95
3.97
7.89 d
in CD3OD 5a-Eu
5a-Zu
in CDCl3 5a-Eu
5a-Zu
2.760 sext bw 24
3.858 t bw 8
2.98v
6.811 d
4.02v
7.315 t
6.92v
7.078 t
7.404v
7.105 t
7.371 t
7.171v
2.728 sext bw 22
3.912 bw 8.4
3.007 bw 22.3
6.789 d
4.052 bw 9.0
7.322 t
6.907 d
8.5 v.br.
7.136 t
7.239 7.1
7.79 br
aThe three methylene group gave complex peaks in the 1.4–2.2 ppm region.
bUnresolved.
c2H.
d1H.
ePart of multiplet 7.32–7.37 ppm.
f6.961 s, 7.448 s ppm CONH2.
g6.87 s, 7.3 s ppm CONH2.
h12.30 ppm COOH.
i12.22 ppm COOH.
j12.15 ppm COOH.
k3-phenyl, similar to data of ref. 9 except for δ1-NH reported as 7.80 ppm.
lJ = 6.6, 7.8 and 9.0 Hz.
mAfter removal of NH coupling by double resonance: J = 6.6, 5.3, and 4.2 Hz.
nPart of m 7.27–7.38 ppm.
o1H under fast exchange; under slow exchange 9.377 ppm 0.4H Z-isomer and 9.146 ppm 0.6H E-isomer.
pImino group proton: under fast exchange 7.40 ppm; under slow exchange 8.59 ppm 0.6H E-isomer, 6.58 ppm 0.4H Z-isomer.
qPart of multiplet 7.39–7.55 ppm.
r2H, common signal for 1-H and the imino proton.
sE:Z:5a = 7.5:1.5:1.
tCould not be assigned.
uE:Z = 1:1.
vBroadened.
In CD3OD only, a 1:1 ratio of the imino Z and E isomers
was observed. The signals of one of the isomers were broad-
ened, which is most likely due to exchange with the amino
form. In CDCl3 only two sets of signals, both sharp, in 1:1
ratio were observed. This could be due either to fast ex-
change with the amino isomer or to its absence in significant
amounts.4
The assignment of Z or E configurations was based on the
expected upfield shift of 4a-H and 7a-H caused by the ring
current effect of the phenyl group, because in the E-isomer
4 One of the referees has suggested that the doubling of the NMR resonances could be due to dimer-polymer equilibria. We believe that the
amidines studied in ca. 0.04 M solutions are unlikely to give associates, particularly in DMSO and MeOH.
© 2000 NRC Canada

Atay et al.
89
Scheme 3.
allylic strain will force it into a perpendicular position. Such
an assignment readily explains the favouring of the E-isomer
in DMSO: solvation at 3-NH will hinder the phenyl group in
the Z-isomer. The conformation, syn or anti, of the 5a iso-
mer could not be assigned.
Table 3. Absorbance ratios in 0.1 M HCl for
determining Ke = [cis-2]/[cis-1] and pKAH of
the phenylthioureido group of nitrile cis-2
(25.0°C and ionic strength of 1 M (KCl)).
[OH–] (mol/dm–3)
A295a/A265
b,c
Noteworthy is the strong bias of cis-2 for conformation A
(Scheme 2) suggested by the bandwidths of the signals for
protons 4a and 7a in DMSO: 22.2 and 11.5 Hz respectively.
7a-H is axial in the pyrimidine ring thus JHNH will be 0–
1 Hz (7) and pseudoequatorial in the cyclopentane ring so
that the sum of 11.5 Hz has to be made up of three gauche
couplings. The signals for the 4a- and 7a-protons in the
imino tautomers of compound 5 were better resolved than
those for the iminopyrimidine 2. Again the bandwidths for
the 7a-H signals are very small: 8.4–9.0 Hz in CD3OD and
CDCl3. The signals are triplets which can be due to two cou-
plings of ca. 4 Hz. This is the value of a cis-coupling in
dihydrouracils in cases of a directed electronegativity effect
(7), i.e., when an electronegative atom is anti to one of the
coupled protons. In conformation A, shown on Scheme 2 for
compound 2, the ring nitrogen is anti to 4a-H and to the
neighbouring cis cyclopentane ring proton which accounts
for the low couplings. The coupling with the trans proton of
the pentane ring with a torsion angle of ca. 90° can be close
to zero and thus not seen. The sextets observed for 4a-H can
be accommodated by two couplings of 9–10 Hz and one of
ca. 4 Hz. The latter is of course J4a,7a while the large cou-
plings are the expected ones with torsions of ca. 30 and 150°
with the cis and trans cyclopentane ring protons, respec-
tively.
0.00200
0.00400
0.00600
0.00800
0.0140
0.0160
0.0200
0.0400
0.100
0.160
0.200
0.320
0.400
1.42
1.43
1.39
1.38
1.32
1.29
1.24
1.05
0.785
0.68
0.634
0.538
0.511
0.502
0.474
0.520
1.00
aA°1,294 = 1 870, A°2,294 = 11 200.
bA°1,265 = 9810, A°2,265 = 4260.
cRatios are averages from several experiments.
Defining Ke = [cis-2]/[cis-1] as the equilibrium constant
between the neutral species, then upon considerable ioniza-
tion of the thioureido group the apparent equilibrium be-
tween the stoichiometric concentrations is given by eq. [1]:
[cis-2]st
[cis-1]st
Ke
[1]
Kapp
=
=
Kinetics of alkaline hydrolysis of nitrile cis-1
The equilibrium nitrile cis-1-imine cis-2
1 + [OH− ]KANH/Kw
The equilibrium mixture was analyzed by quenching the
reaction mixture with hydrochloric acid at the moment of
reaching the maximum of absorption at 267 nm, when the fi-
nal products could still be neglected. HCl was added to a fi-
nal concentration of ca. 0.1 M. Under these conditions the
imine cis-2 was fully protonated (pKBH+ was determined
spectrophotometrically to be 6.40 0.11)5 and hydrolysis is
relatively slow. To avoid errors due to dilution, the equilib-
rium constant (Ke) and the dissociation constant for the
CSNHPh group (KANH) was determined from the ratio of the
absorbances at two wave lengths:
As described above when the hydrolysis of cis-1 in aque-
ous hydroxide is monitored by UV, the peak at 245 nm de-
creases while a new absorption at 267 nm appears due to
cis-2. The latter then breaks down by a much slower process
and the absorption at 245 nm is restored. Quantitative esti-
mates showed that the conversion of cis-1 into cis-2 is in-
complete and an equilibrium is established (Scheme 3). At
an [OH–] higher than 0.01 M the amount of cis-2 at equilib-
rium decreased. This can be explained by considerable ion-
ization of the phenylthioureido group.
5 pK values for 4-imino-3-phenylhydantoins ranging from 6.1 to 6.5 have been reported in ref. 17.
© 2000 NRC Canada

90
Can. J. Chem. Vol. 78, 2000
Fig. 1. Semilogarithmic plot of A294/A265 against [OH–]. Curve is
drawn by means of eq. [2] and parameters in text.
Fig. 2. Determination of pKAH for ionization of the
phenylthioureido group in carboxamide cis-3.
A
A° + A2°1Ke /(1 + KANH[OH]/Kw)
11
11
[2]
=
A
A° + A2°2Ke /(1 + KANH[OH]/Kw)
12
12
where A0ii are the molar absorbances, the first subscript de-
noting the compound and the second one the wave length. A
curve fit of the ratios A294/A265 (Table 3) to eq. [2] in the in-
terval 0.002 to 1 M KOH gave Ke = 2.43 0.11 and KANH /Kw
= 18.0 2.1 (I = 1 M (KCl)). Taking Kw = 1 × 10–14 and
[OH–] = aOH gave a pKANH of 12.74. These parameters re-
sulted the fit shown in Fig. 1. The positive deviation of the
points in 0.5 and 1 M KOH could be due to deprotonation of
the amidine 2 (18) but attempts to include such an ionization
in eq. [2] were unsuccessful.
Due to the rapid conversion of the nitrile cis-1 in the re-
gion of ionization its pK could not be determined independ-
ently by spectroscopy. To check whether the value was
reasonable, the pK for the ionization of the phenylthioureido
group of the related amide cis-3 was determined in KOH
solutions (I = 1 M (KCl)). Again using –log [KOH] as the
acidity function and pKw = 14, a pKAH of 12.78 0.02 was
obtained (Fig. 2).
[3]
kobs = kf [OH− ]/(1 + KANH[OH− ]/Kw) + kf [OH− ]/Ke
The experimental data were fitted eq. [3] using the ioniza-
tion and equilibrium constants found above as fixed parame-
ters. This afforded kf = 4.24 0.08 dm3 mol–1 s–1 (Fig. 3). A
krvalue of 1.74 dm3 mol–1 s–1 was calculated from the equi-
librium constant.
The kinetics observed indicate that the reactive form of
cis-2 is the ureide anion. According to the principle of
microscopic reversibility the return path will involve the
amidine anion.
The breakdown of the iminopyrimidine cis-2 to products
was found to be independent of the concentration of OH– be-
tween 0.001 and 0.012 M KOH, where the observed first-
order rates averaged to 2.21 × 10–4 s–1. The somewhat higher
rate observed in 0.1 M KOH indicates that a first order in
[OH–] process sets in. Actually, taking into account the
lower fraction of cis-2 at equilibrium in 0.1 M KOH, a lower
pseudo-first-order constant should have been observed if
only the pH-independent reaction took place. pH-
independent rates are often encountered in the hydrolysis of
amidines (18) in the alkaline region. The reaction can be for-
mulated as attack of water on neutral iminopyrimidine, how-
ever, attack of OH– on the protonated form as shown in
Scheme 3 appears more appealing. The tetrahedral interme-
diate could either lose ammonia to give the pyrimidine 7
which would rapidly hydrolyze6 to the acid 4, or open the
cycle to form the amide 3 which would then hydrolyze to
the same acid 4.7 1H NMR of extracts of the acidified reac-
tion mixture leaving cis-2 as salt in the acid solution showed
that the second route through the amide is preferred. After
110 min in 0.01 M KOH at room temperature (around two
half-lives of cis-2) nitrile cis-1 gave a mixture of cis-3:cis-
Rates of formation of imine cis-2 and its breakdown
The large difference in the rates of formation of cis-2
from cis-1 and the breakdown of the former allowed these to
be measured as two separate first-order processes. The for-
mation and breakdown were measured only up to 0.04 M
KOH because of the considerable drop (due to the ionization
of the ureido group at the higher alkalinities) in the amount
of intermediate 2 formed at equilibrium. Furthermore, the
rate of formation became too fast to measure by conven-
tional means.
The cyclization of ureido nitrile 1 was found to be first or-
der in [OH–] at lower alkalinities. Taking into account the
equilibria in Scheme 3, the following expression for the de-
pendence of kobs on [OH–] can be derived:
6 At pH 9.18 cis-7 hydrolyzes with an observed rate of 0.0015 s–1.
7 Thin layer chromatography showed general retention of configuration. Only trace trans isomers of the products were found.
© 2000 NRC Canada

Atay et al.
91
Scheme 4.
lyzed elimination giving the thioisocyanate shown on
Scheme 4. The first reaction involves charge separation in
the transition state while the second one dispersal of charge.
Decreasing the dielectric constant by addition of acetonitrile
will favour the second and disfavour the former reaction.
Addition of a polar aprotic solvent will raise specifically the
activity of the ground state of the elimination reaction (OH–)
and thus further accelerate elimination over hydrolysis.
Fig. 3. Pseudo first order rate constants for the cyclization of
nitrile 1. Curve calculated by means of eq. [3] with parameters
in text.
Experimental
Unless otherwise stated IR-spectra are in nujol mulls on a
Bruker IFS instrument, frequencies are in cm–1, wavelengths
are in nm. Instruments included a Bruker Spectrospin WM
250 NMR spectrometer (chemical shifts in ppm against
TMS, couplings in Hz), a Jeol-D 300 mass spectrometer and
a Unicam SP 800 UV spectrophotometer. Uncorrected melt-
ing points were determined in capillaries.
Inorganic reagents for kinetic measurements were of ana-
lytical grade and were used without further purification.
cis- and trans-2-(-3-Phenylthioureido)cyclopentanecarboni-
trile (1): The parent 2-aminocyclopentanecarbonitrile was
prepared as a crude oil as follows. A solution of 2-amino-1-
cyanocyclopentene (20) (7.3 g, 68 mmol) in methanol
(100 mL) was made acid to bromocresol green with 1 N
methanolic HCl. Sodium cyanoborohydride (4.9 g) was
added by portions while the solution was kept acidic by the
addition of more methanolic HCl. After stirring for 3 h the
reaction mixture was concentrated. The residue was treated
with 10% aqueous NaOH (68 mL) and NaCl (14 g), and
taken up in dichloromethane (4 × 50 mL). It was removed
from dichloromethane with 10% HCl (4 × 50 mL). The
aqueous extract was made basic with 20% aqueous NaOH
and extracted with dichloromethane (4 × 50 mL). The di-
chloromethane solution was dried (Na2SO4) and concentrated
under reduced pressure to yield 2-aminocyclopentanecarbon-
4:cis-1 in a ratio of 57:30:13. After 7 h [cis-3]:[cis-4] =
19:81 and in 24 h the extract was pure acid cis-4.
In the introduction the magnitude of the anchimeric assis-
tance to the alkaline hydrolysis of the nitrile group by the
neighbouring phenylthioureido one is quoted as 1 × 104.
This is estimated by comparing the rate of breakdown of cis-
2 with the second order rate constant of alkaline hydrolysis
of acetonitrile. Another comparison could be provided by
the hydrolysis of the nitrile group in trans-1 which has a
closer similarity with respect to steric and electronic influ-
ences. For 18 days at room temperature in aqueous 0.01 M
KOH trans-1 degraded only around 50%, the products were
not, however, the amide trans-3 or the acid trans-4 (1H
NMR), and probably result from transformations of the
thioureido group. If that time is compared with the half-life
of cis-2 of 50 min, the trans nitrile is at least 500 times more
stable towards alkali.
itrile (6.53 g, 88%) as a colourless oil. IR (Nujol) νmax
3370, 3305 (NH), 2220 (CN) 1600 (NH2).
:
A solution of phenylisothiocyanate (2.80 g, 21 mmol) in
ether (5 mL) was slowly added to a solution of the above 2-
aminocyclopentanecarbonitrile (2.2 g, 20 mmol) in ether
(5 mL). A crystalline precipitate formed which was filtered
and washed with petroleum ether to give the crude cis and
trans isomers of 1 (4.4 g, 88%). The crude mixture (2 g) was
chromatographed on 60 g “Florosil” with chloroform as the
eluent. 50 mL fractions were taken and fractions 4 to 7
yielded the more mobile cis isomer while 9 to 25 yielded the
Solvent effect on the breakdown of imine cis-2
The most likely transition states for the reaction of cis-2
in water and in a mixture of water:acetonitrile (1:2 v/v) are
shown on Schemes 3 and 4. The zero order in [OH–] hydro-
lysis of 2 proceeds as suggested above as an attack by OH–
on the protonated form of the imine. On the other hand, the
slow step in the isomerization is usually (19) a base cata-
© 2000 NRC Canada

92
Can. J. Chem. Vol. 78, 2000
trans isomer. Crystallization from ethanol-water (2:1)
yielded 0.67 g solid cis-1 (mp 163–164°C). λ max (H2O):
245; IR (KBr disk) νmax: 3230 (NH), 2232 (CN) 1546
(NHCSNHPh); MS m/z: 245. Anal. calcd. for C13H15N3S: C
63.6, H 6.2; found: C 63.3, H 6.15. trans-1 (1.07 g) had a
mp 156–157°C. λ max (H2O): 244; IR (KBr disk) νmax: 3295,
3181 (NH), 2240 (CN); MS m/z: 245. Anal. calcd. for
C13H15N3S: as above; found: C 63.25; H 6.5.
(mp 152–153°C, (chloroform-petroleum ether), resolidifies
and melts again at 230°C). λ max (H2O): 245; IR (Nujol)
ν
max: 3344, 3167 (NH), 1686 (CO), 1536 (NHCSNHPh). MS
m/z: 264. Anal. calcd. for C13H16N2O2S: C 59.1; H 6.1;
found: C 58.9; H 6.3.
cis-4-Imino-1,2,3,4a,5,6,7,7a-octahydro-3-phenyl-2-thiooxo-
cyclopenta[d]pyrimidine (cis-2): The cis nitrile 1 (36 mg,
0.146 mmol) was dissolved by warming in acetonitrile
(2 mL) and subsequently mixed with 0.01 M KOH (100 mL)
at room temperature. After 3 min the mixture was trans-
ferred to a cooled separatory funnel containing ice and HCl,
the amount of the latter dependant on the desired final acid
concentration. The solution was extracted rapidly with di-
chloromethane (3 × 40 mL) in the course of five to eight min
and passed through a fluted filter. The dry residues of the ex-
tracts amounted to 30–35% (by weight) of the starting mate-
cis-2-(3-Phenylthioureido)cyclopentanecarboxamide (cis-3):
The cis nitrile 1 (0.30 g, 1.2 mmol) was stirred in 60% sul-
furic acid (5 mL). When solution was complete (1 h) it was
diluted with water (15 mL) under cooling. The mixture was
extracted with chloroform and the chloroform layer washed
with 5% sodium carbonate and water, dried, and evaporated
to dryness. TLC of the solid residue of cis amide 3 showed
no appreciable admixture of the trans amide. Recrystalli-
zation from CHCl3–petroleum ether yielded 0.15 g, 47%
(mp 160–161°C after second crystallization). λ max (H2O):
245. IR (KBr disk) νmax: 3388, 3356, 3277, 3165 (NH), 1662
(CONH2), 1541 (NHCSNHPh). MS m/z: 263. Anal. calcd.
for C13H17N3OS: N 15.96; found N 15.54.
1
rial and consisted of (according to H NMR) mixtures of the
initial nitrile and the cis amide, the amount of the latter
varying from trace to 30%). The aqueous layer was made al-
kaline (ca. 0.01 M in [OH–]) and extracted twice rapidly
with chloroform (ca. 12 min from the start of the experi-
ment). The organic layer was filtered, dried with MgSO4,
and evaporated to give a solid, the iminopyrimidine cis-2
(25 mg, containing up to ca. 15% the cis nitrile according to
1H NMR). λ max (H2O): 267, (0.1 M HCl 295); IR (CHCl3)
trans-2-(3-Phenylthioureido)cyclopentanecarboxamide (trans-3):
The same procedure with trans nitrile 1 (0.30 g, 1.2 mmol)
in which the dilution with water followed 5 min of stirring
afforded, after two recrystallizations from chloroform-
petroleum ether, 0.16 g (50%) of trans-3 (mp 155–156°C,
the crude product contained considerable amounts of the cis
amide-3 (determined by TLC)). λ max (H2O): 245. IR (KBr
disk) νmax: 3346, 3209 (NH), 1660, 1633 (CONH2), 1541
(NHCSNHPh). Anal. calcd. for C13H17N3OS: N 15.96;
found N 15.95.
νmax
: 3409, 3288, 3199b (NH), 1647 (C=N), 1528
(NCSNH).
cis-4-Phenylimino-1,2,3,4a,5,6,7,7a-octahydro-3-phenyl-2-
thiooxocyclopenta[d]pyrimidine (cis-5): A solution of 0.5 g
(2 mmol) of cis-2-(3-phenylthioureido)cyclopentyl cyanide
in 80 mL of absolute methanol was refluxed for 3 h. The
solvent was removed by a rotatory evaporator and the solid
residue recrystallized from MeOH–MeCN to give 0.39 g
(78%) of solid cis-5 (mp 231–2°C). MS m/e: M+ 245; λ max
(H2O): 269, 242; IR (KBr disk) νmax: 1670 (C=N),
1564 (δ NH), 1346 (NHCSNH), 3170, 3460 (NH); IR
(CHCl3) νmax: 1664 (C=N) 1535 (δ NH), 1348 (NHCSNH),
3226, 3408 (NH). Anal. calcd. for C13H15N3S: C 63.63, H
6.16, N 17.13, S 13.07; found: C 63.78, H 6.45, N 16.85, S
13.18.
cis-4-Oxo-1,2,3,4a,5,6,7,7a-octahydro-3-phenyl-2-thiooxo-
cyclopenta[d]pyrimidine (cis-7): (a) cis Nitrile 1 (0.305 g,
1.24 mmol) in 1:2 HCl (30 mL) was refluxed for 30 min.
Cooling and filtering afforded 0.289 g (94%) of the crude
thiouracil 7 (mp 267–269°C, ethanol, lit. mp 285–287°C
(5)). λ max (H2O): 278; IR (Nujol) νmax: 3171 (NH), 1711
(CO), 1564 (NHCSNPh); MS m/z: 246. Anal. calcd. for
C13H14N2OS: C 63.4; H 5.7; found: C 63.6; H 5.8. (b)
Refluxing the trans nitrile 1 (60 mg, 0.24 mmol) in 1:2 HCl
(30 mL) for 1 h, cooling, and standing overnight gave as
cis-1,2,3,4a,5,6,7,7a-Octahydro-4-oxo-2-thiooxocyclopenta[d]-
pyrimidine (cis-8): (a) From phenyliminopyrimidine cis-5.
The phenyliminopyrimidine cis-5 (0.37 g, 1.5 mmol) was
dissolved with gentle heating in 30 mL of MeOH and 10 mL
of MeCN. 40 mL of distilled water and 15 mL of 1M HCl
were further added. The solution was left at ambient temper-
ature for 75 min and then evaporated to dryness until HCl
was completely removed in vacuo. Recrystallization from
MeOH–petroleum ether afforded 0.180 g (80%) of solid (mp
198–9°C, lit. mp 206–208°C(11)); MS m/e: M+ 170; IR
(KBr disk) νmax: 1693 (CO), 1362 (NHCSNH), 3182, 3460w
(NH). Anal. calcd. for C7H10N2OS: C 49.39; H 5.92; N
16.46; S 18.91; found C 49.50; H 6.03; N 16.77; S 18.91.
(b) From cyclization of cis-2-thioureidocyclopentanecar-
boxylic acid (cis-6). The acid cis-6 (0.150 g) was dissolved
in 8 mL of MeCN and 4 mL of MeOH. 12 mL of 1 M HCl
were added and the mixture left at room temperature for
20 h. The solvent removed and the dry residue recrystallized
from MeCN and petroleum ether to yield 0.085 g (63%) of
1
precipitate 47 mg (78%) of pure (determined by H NMR)
cis pyrimidine 7.
trans-2-(3-Phenylthioureido)cyclopentanecarboxylic acid (trans-4):
Refluxing the nitrile trans-1 (52 mg, 0.21 mmol) in 1:2 HCl
for 30 min after cooling and standing overnight afforded
30 mg of a 4:1 mixture (1H NMR) of the cis pyrimidine 7
and the trans acid 4. Evaporating the mother liquor with a
rotary evaporator yielded 19 mg, 34% of pure trans acid (mp
146–147°C, lit. mp 142–144°C(5)); λ max (H2O): 245; IR
(KBr disk) νmax: 3346, 3169 (NH), 1701, 1687 (CO), 1549
(NHCSNHPh). Anal. calcd. for C13H16N2O2S: C 59.1; H
6.1; N 10.6; found: C 59.0; H 6.32; N 10.8.
cis-2-(3-Phenylthioureido)cyclopentanecarboxylic acid (cis-
4): The cis pyrimidine 7 (0.5 g, 2.03 mmol) dissolved easily
in warm 0.5 M NaOH (20 mL). After acidifying the cooled
solution with conc. HCl (2 mL) the cis acid 4 separated as a
gum which crystallized rapidly to give 0.53 g. 99% of solid
© 2000 NRC Canada

Atay et al.
93
Table 4. Pseudo-first-order rate constants (s–1) for the formation
and breakdown of 4-imino-2-thiooxo-1,2,3,4a,5,6,7,7a-
octahydrocyclopenta[d]pyrimidine cis-2 in aqueous KOH and
ionic strength 1.0 M (KCl) at 25°C.
jecting 20 to 80 µL of a 10–2 to 10–3 M stock solution of the
substrate in MeCN, to the preheated KOH solution of ionic
strength 1 M (KCl). The conversion of cis-1 into cis-2 was
monitored by the increase of absorption at 267 nm while the
degradation of the latter was monitored by the decrease at
[OH–]
103 kobs for
formation
104 kobs for
breakdown
the same wavelength. Pseudo-first-order rate constants (kobs
)
(mol/dm–3)
were calculated by the least squares procedure from plots of
either ln (At – Ainf) or ln (Ainf – At) (as appropriate) versus
time, where Ainf was the absorbance after ten half-lives.
0.00031a
0.001
0.002
0.004
0.006
0.0064
0.007
0.0074
0.008
0.01
0.012
0.014
0.016
0.0174
0.02
2.48
8.07
12.3
25.8
35.2
43.4
48.2
52.3
44.0
54.0
68.0
72.2
83.9
92.0
87.1
107
2.00
2.24
2.23
2.23
2.22
2.21
2.11
2.17
2.03
2.17
2.48
—
—
—
—
—
—
—
Equilibrium between the cis nitrile 1 and cis
iminopyrimidine 2
Under the conditions of the rate experiments a solution of
the cis nitrile 1 in MeCN was added to the preheated at
25.0°C KOH solution so that the final concentration of 2
was 7–8 × 10–5 M. When maximum absorption at 267 nm
(λ max (H2O) of cis-2) was reached, the solution in the cell
was acidified with a predetermined amount of HCl so that its
final concentration was ca. 0.1 M and the spectrum of the
solution taken immediately. The ratio [cis-2]:[cis-1] was de-
termined from the ratio of the absorbances at the two wave
lengths as well as the known extinction coefficients at these
wavelengths of the two compounds in HCl solutions.
Making the solution alkaline after the measurement restored
the pre-acidification spectrum by better than 5%.
0.024
0.032
0.04
141
164
0.1
—
3.90
aExperiment conducted at pH 10.49 in carbonate buffer 70% base.
Product analysis
The course of the alkaline hydrolysis of cis-1 was also
1
followed by H NMR in order to identify the products. For
these large scale experiments the substrate was dissolved in
an appropriate amount of MeCN so that when added to the
KOH solution, the final concentration of MeCN equalled
that in the kinetic experiment. After being kept at 25°C for
the appropriate period of time, aliquots were acidified and
extracted with dichloromethane, the solvent removed and the
solid cis-8 (mp 197–199°C) identical with the above product
according to spectral data and mixed mp.
cis-2-Thioureidocyclopentanecarboxylic acid (cis-6): (a)
0.31 g of phenyliminopyrimidine cis-5 were dissolved by
gentle heating in 20 mL of MeOH. 20 mL of aqueous 1 M
KOH were added and the solution left at room temperature
for 12 h. Most of the MeOH in the solution was removed on
a rotary evaporator and the alkaline solution extracted twice
with 20 mL portions of ether. The organic layers were dried
and the solvent removed to yield an reddish-brown oil iden-
tified as aniline: M+ 93; same IR and UV spectra as that of
an authentic sample. The aqueous layer was neutralized with
2 equivalents of cationite Dowex 50x8 in H+ form. The solu-
tion (pH two to three) was evaporated and the residue
recrystallized twice from MeCN with addition of petroleum
ether to yield 0.045 g (19%) of cis-6 (mp 155–157°C). MS
m/e: M+ 188. λ max (H2O): 230 (log ⑀ = 3.9), 280 (log ⑀ =
3.1). IR (KBr disk) νmax: 1635, 1560 (delta NH) 1703
(COOH), 3186, 3300, 3350 (NH,OH). Solution resulted a
single spot on TLC and no admixtures were observable in
1
composition determined by H NMR.
Determination of pK-values
(a) cis-Amide 3. A solution (20 µL) of the amide (0.01
M) in MeCN was added to a 25.0°C (2.7 cm–1) preheated
solution of KOH (0.002 to 1 M, I = 1 M (KCl)) in a UV cell,
and the absorbance measured at 255 nm. The pK was ob-
tained by a nonlinear regression fit of the equation
K = (AAH – A)[H+]/(A – AA− )
where A is the absorbance, AAH and AA− the limiting absorb-
ances at low and high basicity respectively. [H+] was as-
sumed equal to antilog(14 – log [OH–]).
1
the H NMR spectrum. (b) 0.095 g of cis-8 were dissolved
in five mL of MeOH by gently heating and 10 mL of 1 M
aqueous KOH added. Similar work up yielded 0.097 g of
raw cis-6 identical (according to H NMR) to the product
(b) cis-Iminopyrimidine 2. The compound was isolated as
a solid as described above and as a solution prepared in
1
MeOH which was kept cooled in ice/water. Aliquots (10 L)
µ
were injected into UV cells containing 25.0°C preheated
0.1 M buffer solutions (formate, acetate, and phosphate)
with I = 1 M (KCl). The absorbances were measured at
295 nm and the pK was calculated as above using the pH-
values measured by means of a Radiometer pHM 84 Re-
search pH-meter with a GK 2401 electrode.
obtained in (a).
Kinetic measurements
The rates were measured under pseudo first order condi-
tions in the thermostated cell compartment of a Unicam SP
800 spectrophotometer. The reactions were initiated by in-
© 2000 NRC Canada

94
Can. J. Chem. Vol. 78, 2000
8. B. Kurtev, M. Lyapova, N. Berova, I. Pojarlieff, A. Orahovatz,
P. Petrova, and N. Mollov. Comm. Dept. Chem. (Bulg. Acad.
Sci.) 1 (1), 51 (1967).
9. E.L. Eliel, S.H. Wilen, and L.N. Mander. Stereochemistry of
organic compounds. Wiley, New York. 1994. p. 773.
10. S. Furberg and L.H. Jensen. J. Am. Chem. Soc. 90, 470 (1968);
C.P. Rohrer and M. Sundaralingam. J. Chem. Soc. Chem.
Commun. 746 (1968).
Acknowledgements
We thank the National Foundation for Scientific Research
of Bulgaria for funding this research and Dr. N. Vasilev for
taking variable temperature NMR spectra.
References
11. P. Sohár, Z. Szöke-Molnár, G. Stájer, and G. Bernáth. Magn.
Reson. Chem. 27, 959 (1989).
12. H.O. Kalinowski, S. Berger, and S. Braun. 13C NMR
Spektroskopie, G. Thieme Verlag, Stuttgart, 1984. Chap. 3.
13. C.L. Perrin. In The chemistry of amidines and imidates. Vol. 2.
Edited by S. Patai and Z. Rappoport. Wiley, Chichester. 1991,
p. 165.
1. (a) E.N. Zil’berman. In Reaktzii nitrilov, Khimiya, Moscow,
1972. Chap. 9; A.I. Mayers and J.C. Sircar. In The chemistry
of the Cyano Group. Edited by S. Patai and Z. Rappoport.
John Wiley, Chichester. 1970. pp. 341–421; (b) J.M. Balquist
and F. I. Goetz. J. Heterocycl. Chem. 9, 937 (1972).
2. K.W. Breukink and R.E. Verkade. Recl. Trav. Chim. Pays-Bas.
79, 443 (1960).
14. L.D. Colebrook, H.G. Giles, and A. Rosowski. Tetrahedron
Lett. 5239 (1972).
3. (a) E.C. Taylor and R.V. Ravindranathan. J. Org. Chem. 27,
2622 (1962); (b) T. Kinoshita, H. Wanatabe, S. Sato, and C.
Tamura. Bull. Chem. Soc. Jpn. 53, 442 (1980); (c) K. Takagi,
H. Morita, and T. Aotsuka. Heterocycles, 26, 175 (1987).
4. S. Wideqvist. Ark. Kemi. 10, 265 (1956).
15. L.I. Spiessens and M.J.O. Anteneus. Bull. Soc. Chim. Belg.
91, 763 (1982).
16. G. Barret and A. Foucaud. Bull. Soc. Chim. Fr. 4594 (1968).
17. V. Machacek, G. Svoboda, and V. Sterba. Coll. Czech. Chem.
Commun. 52, 140 (1987).
18. R.H. de Wolfe. In The chemistry of amidines and imidates.
Edited by S. Patai. Wiley, Chichester. 1975, pp. 349–384.
19. A. Williams and K.T. Douglas. Chem. Rev. 75, 627 (1975).
20. H.E. Schroeder and G.W. Rigby. J. Am. Chem. Soc. 71, 2205
(1949).
5. F. Fülöp, Z. Szakonyi, and G. Bernáth. J. Heterocycl. Chem.
34, 1211 (1997).
6. G. Stájer, Z. Szöke-Molnár, G. Bernáth, and P. Sohár. Tetrahe-
dron, 46, 1943 (1990).
7. A.R. Katritzky, M.R. Nesbit, B.J. Kurtev, M. Lyapova, and
I.G.Pojarlieff. Tetrahedron, 25, 3807 (1969).
© 2000 NRC Canada