Suggested improvement in the ing-manske procedure and Gabriel synthesis of primary amines: Kinetic study on alkaline hydrolysis of N-phthaloylglycine and acid hydrolysis of N-(o-carboxybenzoyl)glycine in aqueous organic solvents

Article abstract of DOI:10.1021/jo961172a

A slight modification of the Gabriel synthesis of primary amines is suggested on the basis of the observed and reported values of rate constants for the alkaline and acid hydrolyses of phthalimide, phthalamic acid, benzamide, and their N-substituted derivatives. The suggested procedure requires shorter reactions time and milder acid-base reaction conditions compared with the conventional acid-base hydrolysis in the Gabriel synthesis. A slight modification in the Ing-Manske procedure is also suggested. Pseudo-first-order rate constants, k_obs, for hydrolysis of N-phthaloylglycine, NPG, decrease from 24.1 × 10^-3 to 7.72 × 10^-3 and 6.12 × 10^-3 s^-1 with increasing acetonitrile and 1,4-dioxan contents, respectively, from 2 to 50% v/v (all the percentages given in the paper are vol %), while increasing the organic cosolvents content from 50 to 80% increases ko_bs from 7.72 × 10^-3 to 19.7 × 10^-3 s^-1 for acetonitrile and from 6.12 × 10^-3 to 52.8 × 10^-3 s^-1 for 1,4-dioxan, in aqueous organic solvents containing 0.004 M NaOH at 35 °C. The rate constants for NPG hydrolysis decrease from 2.11 × 10^-2 to 1.19 × 10^-4 s^-1 with increasing MeOH content from 2 to 84%, in aqueous organic solvents containing 2% MeCN and 0.004 M NaOH at 35 °C.

Full text of DOI:10.1021/jo961172a

J . Org. Chem. 1996, 61, 8063-8068
8063
Su ggested Im p r ovem en t in th e In g-Ma n sk e P r oced u r e a n d
Ga br iel Syn th esis of P r im a r y Am in es: Kin etic Stu d y on Alk a lin e
Hyd r olysis of N-P h th a loylglycin e a n d Acid Hyd r olysis of
N-(o-Ca r boxyben zoyl)glycin e in Aqu eou s Or ga n ic Solven ts
M. Niyaz Khan
Department of Chemistry, Faculty of Science, Universiti Malaya, 50603 Kuala Lumpur, Malaysia
Received J une 21, 1996X
A slight modification of the Gabriel synthesis of primary amines is suggested on the basis of the
observed and reported values of rate constants for the alkaline and acid hydrolyses of phthalimide,
phthalamic acid, benzamide, and their N-substituted derivatives. The suggested procedure requires
shorter reactions time and milder acid-base reaction conditions compared with the conventional
acid-base hydrolysis in the Gabriel synthesis. A slight modification in the Ing-Manske procedure
is also suggested. Pseudo-first-order rate constants, kobs, for hydrolysis of N-phthaloylglycine, NPG,
decrease from 24.1 × 10-3 to 7.72 × 10-3 and 6.12 × 10-3 s-1 with increasing acetonitrile and
1,4-dioxan contents, respectively, from 2 to 50% v/v (all the percentages given in the paper are vol
%), while increasing the organic cosolvents content from 50 to 80% increases kobs from 7.72 × 10-3
to 19.7 × 10-3 s-1 for acetonitrile and from 6.12 × 10-3 to 52.8 × 10-3 s-1 for 1,4-dioxan, in aqueous
organic solvents containing 0.004 M NaOH at 35 °C. The rate constants for NPG hydrolysis decrease
from 2.11 × 10-2 to 1.19 × 10-4 s-1 with increasing MeOH content from 2 to 84%, in aqueous
organic solvents containing 2% MeCN and 0.004 M NaOH at 35 °C.
Sch em e 1
In tr od u ction
The Gabriel synthesis, for converting halides to pri-
mary amines, involves two major steps: (i) the synthesis
of N-substituted phthalimides from halides and (ii) the
synthesis of primary amines from N-substituted phthal-
imides.1 The formation of a primary amine from the
N-substituted phthalimide involves either hydrolysis,
whether acid- or base-catalyzed, or hydrazinolysis (the
Ing-Manske procedure).2 In an effort to improve the
Gabriel synthesis, Osby and co-workers3 have reported
the conversion of phthalimides to primary amines in an
efficient, two-stage operation using NaBH4/2-propanol
and then acetic acid. Several alternatives to the Gabriel
synthesis have been recently reported.2 None of the
reports on improving the Gabriel synthesis appear to be
based on the mechanistic details of the reactions involved.
We wish to report an improvement in the second step of
the Gabriel synthesis based on the detailed mechanisms
of the reactions. The suggested improvement in the
second step of the Gabriel synthesis is supported by the
results of a kinetic study on the alkaline hydrolysis of
NPG and the acid hydrolysis of N-(o-carboxybenzoyl)-
glycine in water-organic solvents. The kinetic results
obtained and their probable explanations are described
in this paper.
Sch em e 2
The complete degradation of phthalimide (PTH) and
its N-substituted derivatives (NSPTH) under basic and
acidic conditions may be represented by Schemes 1 and
2, respectively. The values of the hydroxide ion-catalyzed
Table 1. The literature values of the second-order rate
constants, k1 () k1 /[-OH]) for the hydroxide ion-
catalyzed cleavage of benzamide, N-methylbenzamide,
second-order rate constants, k1
() k1 /[-OH], where
2OH
2
1OH
1
k1 represents the pseudo-first-order rate constant), for
1
the conversion of NSPTH to NSP- are summarized in
and o-methoxybenzamide are also listed in Table 1. The
value of k1
() k1 /[-OH]) for the conversion of NSP-
2OH
2
to phthalate ion (1-) may be slightly smaller than k1
X Abstract published in Advance ACS Abstracts, October 15, 1996.
(1) Gabson, M. S.; Bradshaw, R. W. Angew. Chem., Int. Ed. Engl.
1968, 7, 919; Angew. Chem. 1968, 80, 986.
(2) March, J . Advanced Organic Chemistry: Reactions, Mechanisms
and Structures, 4th ed.; J ohn Wiley & Sons: New York, 1992; p 426
and references cited therein.
2OH
for benzamide because of steric and electrostatic effects.
The literature values of the pseudo-first-order rate
constants, k2 () k21H[H+]), for the H+-catalyzed conver-
1
sion of PTH to phthalamic acid under different conditions
are listed in Table 1. The values of k21 for NSPTH, where
(3) Osby, J . O.; Martin, M. G.; Ganem, B. Tetrahedron Lett. 1984,
2093.
S0022-3263(96)01172-3 CCC: $12.00 © 1996 American Chemical Society

8064 J . Org. Chem., Vol. 61, No. 23, 1996
Khan
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n ts (k) a n d -OH- a n d H+-Ca ta lyzed Secon d -Or d er Ra te Con sta n ts (kOH a n d kH
)
for th e Clea va ge of Su bstr a tes (Am id es a n d Im id es)a
substrate
phthalimide
rate constant
pH, [H+], or [-OH]
T (°C)
105kH ) 9.8 M-1 s-1
10-3kOH ) 4.1 M-1 s-1
kOH ) 26.3 M-1 s-1
kOH ) 31.3 M-1 s-1
kOH ) 33.5 M-1 s-1
kOH ) 2196 M-1 s-1
kOH ) 1.3 M-1 s-1
kOH ) 177 M-1 s-1
109k ) 3.1 s-1
10026
10026
3027
N-(2-bromoethyl)phthalimide
N-(3-bromopropyl)phthalimide
N-(ethoxycarbonyl)phthalimide
N-hydroxyphthalimide anion
N-hydroxyphthalimide
2528
3028
3028
3029
3029
benzamide
0.001 M HCl
1.0 M HCl
4.0 M HCl
4810
104k ) 3.5 s-1
10010
4810
105k ) 1.0 s-1
104kOH ) 15.8 M-1 s-1
104kOH ) 7.2 M-1 s-1
104kOH ) 11.0 M-1 s-1
104k ) 2.4 s-1
100.430
100.430
10031
47.332
47.310
4811
N-methylbenzamide
o-methoxybenzamide
phthalamic acid
pH 1.3-1.8
0.001 M HCl
pH e 2.0
104k ) 2.4 s-1
104k ) 20.0 s-1
N,N-dimethylphthalamic acid
o-HOOCC6H4CONHC6H4X
with X ) H
104k ) 10.0 s-1
g0.002 M HCl
pH ≈ 2
3514
10k ≈ 70.0 s-1
65.815a
24.615a
50.1511
65.815a
50.1511
50.1511
65.815a
50.1511
104k ) 1.9 s-1
0.1 M HCl
0.01 M HCl
pH ≈ 2
104k ) 28.2 s-1
X ) 4′-OMe
) 3′-OMe
) 3′-Me
104k ≈ 117.0 s-1
104k ) 24.1 s-1
0.01 M HCl
0.01 M HCl
pH ≈ 2
104k ) 36.2 s-1
X ) 3′-Cl
104k ≈ 25.0 s-1
104k ) 10.4 s-1
0.01 M HCl
a
Reaction solvent is water.
Sch em e 3
R represents alkyl groups, may not be significantly
different from k2 for PTH under essentially similar
1
conditions. The literature values of the first-order rate
constants, k2 , for conversion of phthalamic and a few
2
N-substituted phthalamic acids (NSPH) to phthalic acid
(1) are also summarized in Table 1.
In view of the reported values of k1 , k1 , k2 , and k2 ,
1
2
1
2
it is apparent that for the least reactive NSPTH, the half-
life period, t1/2, for the k1 step is nearly 0.3 s at 30 °C
1
(0.02 s at 60 °C) and 0.1 M NaOH, while t1/2 for k2 is
1
nearly 13 750 h at 30 °C (175 h, at 60 °C) and 0.1 M HCl
under similar experimental conditions. Similarly, t1/2 for
the k1 step is nearly 17 500 h at 30 °C (260 h at 60 °C)
2
and 0.1 M NaOH, while t1/2 for the k2 step is nearly 2 h
2
at 30 °C (3 min at 60 °C) and 0.1 M HCl under essentially
similar other experimental conditions.
Because of the significantly large values of t1/2 for the
k12 step in alkaline medium and for the k21 step in acidic
medium in the complete degradation of NSPTH, the use
of both basic and acidic hydrolysis of NSPTH turned out
to be an inefficient experimental procedure to prepare
primary amines in the Gabriel synthesis. A more useful
and widely used modified procedure of the Gabriel
synthesis is the Inge-Manske procedure, which involves
the hydrazinolysis of NSPTH to generate the desired
primary amines.2 The detailed reaction mechanism for
hydrazinolysis of PTH is shown in Scheme 3.4
The reported value of k1 is 22.4 M-2 s-1 at 30 °C,4 which
shows that t1/2 for the k1 step is nearly 3 s at 30 °C and
0.1 M NH2NH2. The cyclization of 2 to 3 is expected to
be catalyzed by hydroxide ion. The values of the second-
order rate constants for the hydroxide ion-catalyzed
cyclization of phthalamide,5 N′-dimethylphthalamide,5
o-(aminomethyl)benzamide,6 and o-(hydroxymethyl)ben-
zamide7,8 are 4.9 (25.9 °C), 7.6 (25.9 °C), 0.16 (30 °C), 0.04
(25 °C),7 and 0.15 M-1 s-1 (30 °C),8 respectively. In view
of these results, if we assume that the value of k2 is nearly
0.2 M-1 s-1 at 30 °C, then the t1/2 values for the k2 step
at pH 10 are nearly 10 h at 30 °C and 30 min at 60 °C.
It is to be noted that these conclusions are valid for the
reactions carried out in 100% water solvent. The change
in reaction solvent from 100% water to organic or mixed
aqueous organic solvents is expected to reduce the values
of both rate constants, k1 and k2. It is now evident that
(4) Khan, M. N. J . Org. Chem. 1995, 60, 4536.
(5) Shafer, J . A.; Morawetz, H. J . Org. Chem. 1963, 28, 1899.
(6) Fife, T. H.; DeMark, B. R. J . Am. Chem. Soc. 1977, 99, 3075.
(7) Chiong, K. N. G.; Lewis, S. D.; Shafer, J . A. J . Am. Chem. Soc.
1975, 97, 418.
(8) Okuyana, T.; Schmir, G. L. J . Am. Chem. Soc. 1972, 94, 8805.

Improvement in the Ing-Manske Procedure
J . Org. Chem., Vol. 61, No. 23, 1996 8065
the significantly lower value of t1/2 for the k2 step
compared to that for the k1 step or k2 step makes the
2
1
Inge-Manske procedure superior to the usual basic and
acidic hydrolysis of NSPTH (Schemes 1 and 2) involved
in the Gabriel synthesis.
Note: The Inge-Manske procedure may be expected to
be significantly improved in terms of reducing the time
period for generating the primary amine from NSPTH if
the pH of the reaction mixture is increased by adding
NaOH solution to it after completion of the k1 step. The
significant increase in the pH of the reaction mixture at
the start of the k1 step is bound to convert a significant
amount of NSPTH to NSP-, which is relatively much less
reactive toward both -OH and NH2NH2.
F igu r e 1. Plot of absorbance (AU) at 310 nm against time
for a mixed aqueous solution of N-(o-carboxybenzoyl)glycine
(3.57 × 10-3 M) in 71.4% MeCN, 0.108 M HCl at 35 °C.
Exp er im en ta l Section
Ma ter ia ls. Reagent-grade N-phthaloylglycine, NPG, and
phthalimide, PTH, were obtained from Fluka and BDH. All
other chemicals used were also reagent-grade commercial
products. The stock solutions of NPG and PTH were prepared
in acetonitrile.
respective values of k2 and k2/k1 (when A ) N,N-dimeth-
ylphthalamic acid) changed from 1.19 × 10-2 to 1.48 × 10-4
s-1 and 22.5 to 0.1 with increasing MeCN content from 2 to
70% in mixed aqueous solvents containing 0.005 M HCl at 25
°C.13,14 These results show that the presence of PAn in the
hydrolysis of A can be easily detected spectrophotometrically
by using water-acetonitrile or water-aprotic organic solvents
with high contents of organic cosolvent. Phthalic acid and A
do not absorb significantly, while PAn absorbs strongly at 310
nm. Thus, the rates of formation and decay of PAn were
studied spectrophotometrically at 310 nm in H2O-MeCN
solvents containing >60% MeCN.
In a typical kinetic run with a total volume of 5 mL of the
reaction mixture containing 0.004 M NPG, 0.02 M NaOH, and
80% MeCN, the reaction was allowed to complete a period of
more than 50 half-lives (i.e. > 600 s) at 35 °C. The hydrolysis
of the hydrolytic product of NPG (i.e., A in eq 2) was then
initiated by adding 0.6 mL of 1.18 M HCl to the reaction
mixture. The resulting reaction mixture, having a total
volume of 5.6 mL, contained 3.57 × 10-3 M A, 0.108 M HCl,
and 71.4% MeCN. The change in the absorbance, AU, at 310
nm was monitored as a function of time (t) using a diode-array
spectrophotometer. The observed data (for a typical kinetic
run) are shown in Figure 1.
Kin et ic Mea su r em en t s. (a ) Alk a lin e H yd r olysis of
NP G. Under aqueous alkaline and acidic pH conditions, NPG
absorbs strongly (molar absorption coefficient at 300 nm is
nearly 2 × 103 M-1 cm-1) while the hydrolysis product of NPG,
N-(o-carboxybenzoyl)glycine, absorbs weakly at 300 nm (molar
absorption coefficient is < 50 M-1 cm-1). When the water
content was changed from 98 to 15% (all percentages given in
the paper are vol %), mixed aqueous-organic solvents did not
reveal a significant change in the molar absorption coefficients
of NPG and its hydrolysis product at 300 nm. Thus, the rates
of alkaline hydrolysis of NPG were studied spectrophotometri-
cally by monitoring the disappearance of the reactant (NPG)
at 300 nm. The details of the kinetic procedure are described
elsewhere.9
All the kinetic runs were carried out under experimental
conditions in which the reaction rate obeyed pseudo-first-order
kinetics. Pseudo-first-order rate constants, kobs, were calcu-
lated from eq 1
AU ) δapp[X]0 exp(-kobst) + AU∞
(1)
The hydrolysis of phthalamic acid was carried out as follows.
The alkaline hydrolysis of PTH in a reaction mixture (15 mL)
containing 4.4 × 10-3 M PTH, 0.02 M NaOH, and 70% MeCN
was allowed to complete a period of nearly 7 half-lives (i.e.,
nearly 21.5 h) at ambient temperature (≈28 °C). The hydroly-
sis of phthalamic acid was then initiated by adding 0.8 mL of
1.18 M HCl to 5.0 mL of the reaction mixture. The resulting
reaction mixture contained 3.79 × 10-3 M phthalamic acid,
0.146 M HCl, and 60.3% MeCN. The change in absorbance
at 310 nm was monitored as a function of time using a diode-
array spectrophotometer. The observed data (AU versus t) is
shown in Figure 2.
using a nonlinear least-squares technique considering δapp
(apparent molar absorption coefficient) and AU∞ (the absor-
bance at t ) ∞) as unknown parameters. In eq 1, AU is the
absorbance at any reaction time, t, and [X]0 is the initial
concentration of NPG. The reactions were carried out for up
to 4-9 half-lives, and the observed data fitted eq 1 well.
(b) Aqu eou s Clea va ge of N-(o-Ca r boxyben zoyl)glycin e
a n d P h th a la m ic Acid in Acid ic Med iu m . The aqueous
cleavage of phthalamic acid and N-substituted phthalamic
acids (NSPH) at pH e 3 is known to follow an irreversible
consecutive reaction path (eq 2)
k1
k2
Resu lts a n d Discu ssion
A
98 PAn
98 C
(2)
(a ) Effects of Mixed Aqu eou s-Or ga n ic Solven ts on
th e Ra te of Alk a lin e Hyd r olysis of NP G. A series of
kinetic runs was carried out at 0.004 M NaOH and 35
°C in mixed aqueous solvents containing different con-
tents of MeCN (2-84%), 1,4-dioxan (2-83%), and MeOH
(2-84%). A few kinetic runs were also carried out within
the 1,4-dioxan content range 10-80% at 0.008 M NaOH,
2% MeCN, and 35 °C. Pseudo-first-order rate contents,
where A, PAn, and C represent phthalamic acid or NSPH,
phthalic anhydride, and phthalic acid, respectively. It is
interesting to note that in the classic paper on hydrolysis of
phthalamic acid, Bender et al. used an ingenious technique to
affirm indirectly the formation of PAn.10 In 1977, Blackburn
et al.11 showed spectrophotometrically the formation and decay
of PAn in the hydrolysis of phthalamic acid in the presence of
high concentrations of sodium perchlorate, where the rate of
hydrolysis of PAn was greatly retarded.12 We found that the
k
obs, are shown graphically in Figure 3. Increasing MeCN
from 2 to 50% decreased kobs from 2.41 × 10-2 to 7.72 ×
(9) Khan, M. N. J . Org. Chem. 1983, 48, 2046.
(10) Bender, M. L.; Chow, Y.-L.; Chloupek, F. J . Am. Chem. Soc.
1958, 80, 5380.
(11) Blackburn, R. A. M.; Capon, B.; McRitchie, A. C. Bioorg. Chem.
1977, 6, 71.
10-3 s-1, while a further increase from 50 to 84%
increased kobs from 7.72 × 10-3 to 2.63 × 10-2 s-1
.
(12) Bunton, C. A.; Fendler, J . H.; Fuller, N. A.; Perry, S.; Rocek, J .
J . Chem. Soc. 1963, 5361.
(13) Khan, M. N. Indian J . Chem. 1993, 32A, 387.
(14) Khan, M. N. Indian J . Chem. 1993, 32A, 395.

8066 J . Org. Chem., Vol. 61, No. 23, 1996
Khan
Sch em e 4
F igu r e 2. Plot of absorbance (AU) at 310 nm against time
for a mixed aqueous solution of phthalamic acid (3.79 × 10-3
M) in 60.3% MeCN, 0.146 M HCl at 35 °C.
The hydroxide ion catalyzed cleavage of NPG involves
-OH and ionized NPG as the reactants. The rate of
reaction of H2O with ionized NPG is negligible compared
with the rate of reaction of -OH with ionized NPG at
[-OH] g 0.002 M.
It is interesting to note that MeO- reacts with acetyl
salicylate ion,16 AS-, and phenyl acetate17 nearly 60 times
faster than -OH in aqueous methanol solvents. Pseudo-
first-order rate constants for the cleavage of AS- revealed
an increase with increasing MeOH content (within the
range 2-18%) in MeOH-H2O solvents containing 2%
MeCN and 0.01 M NaOH.16 Such an increase could not
be detected in the cleavage of ionized NPG under es-
sentially similar experimental conditions. This may be
explained by the simplified reaction mechanism as shown
in Scheme 4 for the alkaline cleavage of NPG in MeOH-
H2O mixtures.
The value of k1 is 5 M-1 s-1 at 30 °C in an aqueous
4
4
4
4
solvent containing 2% MeCN. If k2 /k1 ≈ 60, then k2
300 M-1 s-1 at 30 °C. The values of the rate constants,
OH, for -OH-catalyzed cyclization of methyl o-carbam-
≈
k
oylbenzoate, methyl o-(aminomethyl)benzoate, and ethyl
o-(hydroxymethyl)benzoate are 3.1 × 103 (25.9 °C),5 7 ×
103 (30 °C),18 and 104 M-1 s-1 (30 °C),19 respectively.
These values of kOH predict a value of k-2 as ≈ 103 s-1
,
4
provided Ka/Kw ≈ 1 M-1, where Ka is the ionization
constants of 8H. This analysis indicates that at [MeO-]
e 0.004 M, the value of k-24/k2 [MeO-] g 800. Thus,
4
under the experimental conditions imposed, the ratio
[8]/[NPG] e 1/800 and hence insignificant amounts of 8
or 8H are expected to exist during the course of the
cleavage of NPG.
The decrease in kobs with increasing organic cosolvent
content (i.e., decreasing dielectric constant, ꢀ, of the
reaction medium) may be explained in terms of simple
electrostatic theory.20 But in the electrostatic treatment
of the effects of ꢀ on reaction rates, the charges on the
reactants are presumed to be at the reaction site, while
in the present system the negative charge on the reactant
NPG is not entirely at the specific reaction site. Thus,
F igu r e 3. Plots of kobs at 35 °C (for alkaline hydrolysis of
N-phthaloylglycine, NPG) against the percentage contents of
organic cosolvent (OCS) of reaction mixtures containing 4 ×
10-4 M NPG, 0.004 M NaOH, and mixed solvents, H2O-MeCN
(_), H2O-1,4-dioxan (~) with 2% MeCN, and (3) with 2%
MeCN and 0.008 M NaOH, and H2O-MeOH (.) with 2%
MeCN.
Similarly, increasing 1,4-dioxan from 10 to 50% decreased
kobs from 1.71 × 10 -2 to 6.12 × 10-3 s-1 at 0.004 M NaOH
and from 4.00 × 10-2 to 1.93 × 10-2 s-1 at 0.008 M NaOH.
But increasing 1,4-dioxan from 50 to 83 (at 0.004 M
NaOH) and 80% (at 0.008 M NaOH) increased kobs from
6.12 × 10-3 to 7.57 × 10-2 s-1 and 1.93 × 10-2 to 0.145
s-1, respectively. Similar observations have been re-
ported in several related reactions.15 The rate constants
decreased from 2.11 × 10-2 to 1.19 × 10-4 s-1 with
increasing MeOH content from 2 to 84% at 0.004 M
NaOH and 2% MeCN.
(15) (a) Hawkins, M. D. J . Chem. Soc., Perkin Trans. 2 1976, 642
and references cited therein. (b) Benko, J .; Holba, V. Collect. Czech.
Chem. Commun. 1978, 43, 193. (c) Khan, M. N. Indian J . Chem. 1986,
25, 831. (d) Khan, M. N.; Abdullahi, M. T.; Mohammad, Y. J . Chem.
Res., Synop. 1990, 52; J . Chem. Res. Miniprint 1990, 473.
(16) Khan, M. N.; Gleen, P. C.; Arifin, Z. Indian J . Chem., in press.
(17) Hupe, D. J .; J encks, W. P. J . Am. Chem. Soc. 1977, 99, 451.
(18) Fife, T. H.; DeMark, B. R. J . Am. Chem. Soc. 1976, 98, 6978.
(19) Fife, T. H.; Benjamin, B. M. J . Am. Chem. Soc. 1973, 95, 2059.

Improvement in the Ing-Manske Procedure
J . Org. Chem., Vol. 61, No. 23, 1996 8067
Ta ble 2. Ca lcu la ted P a r a m eter s fr om Eq 3a
103[X]0
MeCN (%, v/v)
[HCl] (M)
δPAnc (M-1 cm-1
)
105k1 (s-1
)
104k2 (s-1
)
AU0
tfin (s)
b
d
sub
NCG
3.85
3.57
5.36
3.70
5.56
3.79
3.86
67.3
71.4
71.4
74.1
74.1
60.3
70.2
0.026
0.108
0.108
0.069
0.069
0.146
0.024
230
210
210
200
200
255
216
8.47 ( 0.23e
5.72 ( 0.29
8.22 ( 0.20
6.70 ( 0.23
7.57 ( 0.18
17.8 ( 0.4
3.20 ( 0.13e
1.14 ( 0.04
1.62 ( 0.06
1.02 ( 0.02
1.37 ( 0.06
0.021 ( 0.001e
0.034 ( 0.004
0.025 ( 0.002
0.043 ( 0.004
0.032 ( 0.002
0.044 ( 0.004
0.082 ( 0.002
4353
25 103
6573
21 503
7173
PHAf
3.70 ( 0.08 (4.85)g
1.76 ( 0.02 (1.79)h
11 223
14 343
20.2 ( 0.2
a
b
Conditions: Mixed H2O-MeCN solvent, 35 °C, λ ) 310 nm. Initial concentration of substrate (sub). c Molar absorption coefficient
of PAn at 310 nm from ref 13. Final time attained in the kinetic run. e Error limits are standard deviations. f Phthalamic acid. k2
value at 60% MeCN from ref 13. k2 value at 70% MeCN from ref 13.
d
g
h
the use of the simple electrostatic theory to predict
quantitatively the effects of ꢀ on the reaction rates of such
reacting systems is rather hazardous.
An alternative possibility, that the reactant ions
(anionic NPG and -OH) will, on average, be surrounded
by water molecules to a larger extent in water-acetoni-
trile and water-1,4-dioxan than in water-methanol
mixtures, thus increasing the reaction rates in the former
mixed solvents because of a higher dielectric constant in
the immediate environment of the reacting ions, cannot
be ruled out completely. But under such circumstances,
the maximum value of ꢀ in the immediate neighborhood
of the reacting ions cannot be larger than ꢀ of pure water
solvent. Thus, the kobs at 98% H2O must be larger than
The increase in kobs with increasing MeCN and 1,4-
dioxan in mixed aqueous solvents containing water e
50-60% indicates that the effects of ꢀ on kobs become
unimportant under such experimental conditions. It is
interesting to note that although ꢀ of MeCN is not
significantly different from ꢀ of MeOH,21-23 the kobs versus
percent organic cosolvent profile for MeCN is different
from that for MeOH at g50% (Figure 3). Although
several explanations may be given to explain such
observations, we prefer the specific medium effect, which
is the consequence of the preferential solvation of a
specific type of ion (either cation or anion) by the organic
cosolvent in the mixed aqueous solvent as the most
effective solvent effect on the rates of reaction described
in this paper. However, in the mixed aqueous-organic
solvents where the organic cosolvent is worse than water
in solvating ionic solutes, solutes of high charge density
are preferentially solvated only by water molecules, and
therefore, differential solvation effects on reaction rates
may be realized only at very high contents of the poor
solvating component of mixed aqueous solvents. Aprotic
organic cosolvents (such as acetonitrile and 1,4-dioxan)
solvate anionic reactant and transition state very poorly
compared with protic organic cosolvents (such as metha-
nol) in mixed aqueous solvents of similar ꢀ. The cations
are apparently more strongly solvated compared to
anions by aprotic organic cosolvents. Such differential
solvation effects on anions and cations are expected to
either increase the stability of the ion-pair formed
between anion and cation or increase the nucleophilicity
of the anionic nucleophile if ion-pair formation is pre-
vented by geometric constraints.
k
obs obtained at any content of MeCN or 1,4-dioxan. The
larger values of kobs at 84% MeCN (kobs ) 2.63 × 10-2
s-1) and 83% 1,4-dioxan (kobs ) 7.57 × 10-2 s-1) than in
98% H2O (kobs ) 2.41 × 10-2 s-1) under similar experi-
mental conditions make this alternative possibility less
attractive.
(b) Acid ic Aqu eou s Clea va ge of N-(o-Ca r boxyben -
zoyl)glycin e (NCG) a n d P h th a la m ic Acid in Mixed
H2O-MeCN Solven ts. The observed data (absorbance
versus time) shown in Figures 1 and 2 reveal the forma-
tion and decay of an intermediate with the progress of
the reaction. This shows that the rates of hydrolysis of
NCG and phthalamic acid follow the reaction scheme
shown by eq 2. The observed data were treated using
eq 3
δ′[X]0k
AU ) AU0 +
1 [exp(-k1t) - exp(-k2t)]
(3)
(k2 - k1)
where [X]0 is the initial concentration of the reactant
(NCG or phthalamic acid), δ′ ) δPAn - δ, and AU0 ) δ-
[X]0 with δ ) δA ≈ δC. The notations δA, δPAn, and δC
represent the molar absorption coefficients of A, phthalic
anhydride, and phthalic acid, respectively. The assump-
tion that δA ≈ δC at 310 nm is supported by the spectra
of phthalamic acid and phthalic acid.11 The unknown
parameters k1, k2, and AU0 were calculated from eq 3
using a nonlinear least-squares technique with known
values of δ′ and [X]0. Since the molar absorption coef-
ficients of N,N-dimethylphthalamic acid14 and phthalic
acid13 turned out to be nearly zero at 310 nm within
1-80% MeCN content range, we consider δ′ ) δPAn and
the values of δPAn at different % MeCN were obtained
from ref 13.
The nonlinear least-squares calculated values of k1, k2,
and AU0 under different reaction conditions are sum-
marized in Table 2. The values of k2 obtained for the
hydrolysis of NCG and phthalamic acid are comparable
with the first-order rate constants for hydrolysis of
authentic PAn under similar experimental conditions.
The vaules of k1 for phthalamic acid at 60.3 and 70.2%
MeCN and 35 °C may be compared with the reported
values of k1 at 100% H2O and 48 °C (Table 1). The value
The charge density of -OH is apparently larger than
that of anionic NPG. Thus, the solvation shell of anionic
NPG is expected to contain larger number of aprotic
organic cosolvent molecules compared to the solvation
shell of -OH in mixed aqueous-aprotic organic cosolvent
with low water content. This characteristic will cause
stronger ion-pair formation between anionic NPG and
Na+ than that between -OH and Na+, and this in turn
will cause a stronger increase in the electrophilicity of
the carbonyl carbon of anionic NPG and a weaker
decrease in the nucleophilicity of -OH with increasing
MeCN or 1,4-dioxan content. This explains qualitatively
the observed increase in kobs with increasing MeCN and
1,4-dioxan contents (Figure 3).
(20) Frost, A. A.; Pearson, R. G. Kinetics and Mechanisms; Wiley:
New York, 1961; p 150.
(21) Franks, F.; Ives, D. J . G. Quart. Rev. 1966, 20, 1.
(22) Engberts, J . B. F. N. In Water A Comprehensive Treatise;
Franks, F., Eds.; Plenum: New York, 1979; Vol. 6, p 139.
(23) Anantakrishnan, S. V. J . Sci. Indust. Res. 1971, 30, 319.

8068 J . Org. Chem., Vol. 61, No. 23, 1996
Khan
of k1 () 7.5 × 10-5 s-1) for NCG is nearly 2.5 times
smaller than k1 for phthalamic acid. This is reasonable,
since the reported F values for N-arylphthalamic acids
are -1.2311 and -1.05 in pure water solvent and σ*CH2COOH
is larger than σ*H, σ*CH2COOMe ) 0.7 and σ*H ) 0.49.24
Su ggested Exp er im en ta l P r oced u r e for th e Sec-
on d Step in th e Ga br iel Syn th esis. The reaction
mixture containing x mol of NSPTH and (x + 0.01) mol
of NaOH dissolved in an appropriate solvent should be
allowed to reflux until all the NSPTH is converted to the
corresponding anionic phthalamic acid (NSP-). The
reflux period is expected to be dependent on the solvent,
the nature of the NSPTH, and the temperature of the
reaction medium. It may vary from a few seconds to a
few minutes with the change in the solvent from pure
water to the mixed aqueous-organic solvent with organic
cosolvent content of e70%. It may be noted that the
increase in the content of MeCN from 1 to 70% in mixed
H2O-MeCN solvent caused nearly a 12-fold decrease in
the rate of hydrolysis of PTH at 55 °C and 0.01 M KOH.25
But the rate of hydrolysis of NPG increased by nearly
2.5- and 9.0-fold, respectively, with increasing MeCN and
1,4-dioxan contents from 50 to 80% at 35 °C and 0.004
M NaOH. After the completion of the k1 step (Scheme
1
1), (x + 0.02) mol of HCl should be added to the reaction
mixture so that the pH of the reaction mixture becomes
e2. The reaction mixture is refluxed again until the k2
1
step (Scheme 2) is completed. The reflux period may vary
from a few minutes (in pure water solvent) to 1 h or less
(in mixed H2O-MeCN solvent). Pseudo-first-order rate
constants for the conversion of N,N-dimethylphthalamic
acid to phthalic acid at 25 °C and 0.005 M HCl were
found to be unchanged with changing MeCN content from
10 to 90% in mixed H2O-MeCN solvent.14
Ch oice of th e Solven ts. The most suitable solvent
is water. But most NSPTH compounds may be sparingly
soluble in pure water. Mixed aqueous-organic solvents
may be used in order to achieve better solubility of
NSPTH. But the organic cosolvent must be an aprotic
solvent such as acetonitrile, 1,4-dioxan, or tetrahydrofu-
ran. Protic organic solvents such as methanol, ethanol,
etc., should be avoided because of the relatively slow
alkaline hydrolysis of NSPTH in these solvents. If
NSPTH carries an easily ionizable proton that is not
attached to nitrogen, then mixed water-aprotic organic
solvents are even better than water.
(24) Hine, J . Structural Effects on Equilibria in Organic Chemistry;
Wiley: New York, 1975; p 91.
Note: At the end of the reaction in the k2 step (Scheme
(25) Khan, M. N.; Sumaila, M. B. U.; Muhammad, A. M. J . Chem.
Res., Synop. 1991, 233; J . Chem. Res., Miniprint 1991, 2301.
(26) Zerner, B.; Bender, M. L. J . Am. Chem. Soc. 1961, 83, 2267.
(27) Khan, M. N. J . Chem. Soc., Perkin Trans. 2 1990, 435.
(28) Khan, M. N. Int. J . Chem. Kinet. 1987, 19, 143.
(29) Khan, M. N. Int. J . Chem. Kinet. 1991, 23, 567.
(30) Bunton, C. A.; Nayak, B.; O’Connor, C. J . Org. Chem. 1986,
33, 572.
2
2), the organic cosolvent may be separated from the
aqueous layer by salting out.
Ack n ow led gm en t. I thank the University of Ma-
laya for Research Vote F408/96.
(31) Bruice, T. C.; Tanner, D. W. J . Org. Chem. 1965, 30, 1668.
(32) Bender, M. L. J . Am. Chem. Soc. 1957, 79, 1258.
J O961172A