- Email: [email protected]
The Kinetics of the Hydrolysis of Homatropine*
OF THE AMERICAN PHARMACEUTICAL ASSOCIATION JOURNAL
878
from the multiple dose study B than an identical single dose of sustained release powder PD. The results of further studies with other compounds relating in vitro release and blood concentration will be reported. SUMMARY
1. Oral administration of a single 4-Gm. dose of SMTD resulted in rapid absorption of the drug with peak concentrations occurring in approximately one hour. 2. The rate of disappearance of the drug from blood could be approximated as first order during the early time intervals, permitting an estimation of the standard performance indexes k b and ( I ! ’ / & , . 3. For these subjects the average k b equaled 0 33, and ( t 1 / 8 ) b equaled two hours, 6 minutes for free SMTD. 4. When adult subjects received 1 Gm. of SMTD followed by 0.35 Gm. hourly for eight hours, steady state blood concentrations of free drug were maintained near 2-4 mg. yo during the one to eight-hour time interval. Reasonably uniurinary excretion Of from such a dosage regimen during the interval of administration. These data illustrate the use of kb in estimating the rate a t which drug must enter the blood in order to maintain fairly constant concentrations.
VijI,
S L V I I , No. 12
5 . The duration of absorption was approxi. mated in subjects receiving single doses of
‘
sustained suspensions to Gm. of SMTD. In subjects receiving powder PD peak blood concentrations occurred between two and four hours, and absorption was considered essentially complete after about four hours. Blood concentrations in the majority of the subjects receiving sustained release powder SD were more prolonged and the duration of absorption appeared to extend over most of the twelve-hour interval studied. 6. The results indicate a possible relationship between the rate of in vitro release and the duration of blood concentrations in human subjects, and the possible value of utilizing these relationships in formulation of some oral sustained release dosage forms. REFERENCES
(1) Foltz. E. L., Swintosky. J . v., and Robinson, M. J , FedernlroFr Proc.. 15 ( I ) , 422(195(i). ( 2 ) Swintosky, J. v., Robinson, M. J , Foitz. E. L . , and Free, S. M., THIS J O U R N A L , 46, 399(1957). (3) Swintosky V . , Robinson, M . J., and Foltz, E. L., jbi;ij v,, Foltz, E , L,, Bondi, A . , Jr., and
:$:$&:$;:
R o & ; ” ~ ; & ~ ~ ~ +; ,4 $ , , 4 ~ o ~ ~ ? , ( ~ ~Kobinson. ~ ~ , , M, J., ;bid., 7W19583. ( G ) Bratton. A. C . , and Marshall, E. K . , J r . , J . B i d . Chunr., 128, 537(1939). (7) Frisk, A. R . . Actn M e d . S c a n d . S i r p p l C ‘ X L I I , Jan., 19-13.
The Kinetics of the Hydrolysis of Homatropine* By J. L. PATELt and A. P. LEMBERGER An investigation of the chemical kinetics of the hydroxyl ion catalyzed hydrolysis of homatropine has shown that the mechanism of the hydrolytic deterioration consists of two reactions. The heat of activation for the two reactions involving the free and the acid form of the base as calculated from Arrhenius plots are 12.3 and 11.4 kilocalories per mole, respectively. It has been found that homatropine free base degrades at a rate approximately five times that of atropine free base, and the salt form of homatropine degrades at a rate approximately eight times that of the salt form of atropine.
of homatropine in solution is due mainly to the hydrolysis of the ester linkage yielding tropine and mandelic acid. Previous investigations (1-4) have shown that a t low pH homatropine is very stable even a t autoclave temperatures. A t higher pH it deteriorates and the rate of deterioration increases ETERIORATION
* Received April 25, 1958, from the School of Pharmacy, University of Wisconsin, hladison. t Present address: Alembic Chemical Works. Baroda. India. Supported in part by the Research Committee of the Graduate School from funds supplied by the Wisconsin Alumni Research Foundation. This paper is adapted from the prize-winning manuscript submitted by J. L. Patel in the Central Region, 1958. Lunsford Richardson Pharmacy Award competition. Presented to the Scientific Section, A.Pn.A.. Los Angeles meeting. April 1958.
uith temperature and hydroxyl ion concentration. The problem, however, has not been studied from the point of view of chemical kinetics. Many publications are found in the literature concerning the relative rates of second order alkaline hydrolysis of aliphatic carboxylic esters. Ingold ( 5 ) has shown the accelerating effect of C1, OH, COOMe, COMe, and CHzCOMe groups and the retarding effect of a negative ionic charge on the alkaline hydrolysis of aliphatic carboxylic esters. No studies, however, have been made to evaluate quantitatively the influence of substituent and environment upon the alkaline hydrolysis of highly substituted esters.
Decernber 1958
SCIENTIFIC EDITION
To evaluate quantitatively the contribution of polar and stearic factors and the positive ionic charge t o the rate of tropine ester hydrolysis, a study on t h e kinetics of the hydrolysis of homatropine, homatropine methyl bromide, and atropine methyl bromide was undertaken. This report presents a part of this investigation and deals specifically with t h e kinetics of the hydrolysis of homatropine. Another objective of this study was t o establish a basis for predicting the rate of hydrolysis of hornatropine i n solution at various hydroxyl ion concentrations and temperatures.
THEORETICAL CONSIDERATIONS The acid catalyzed hydrolysis of homatropine is slow. A study of 0.02 M homatropine in 0.05 M HC1 a t 25" showed no appearance of additional acid over a period of two days. An hydroxyl ion catalyzed reaction may then be the mechanism responsible for the hydrolysis of homatropine in solution. Homatropine can occur as free base or as the charged ion in solution. Two hydrolytic pathways are therefore possible. On the theoretical concepts employed by Higuchi, et al. (6). in their study on procaine, it is expected that in relatively high pH solutions where hornatropine is mainly present as free base and in low pH solutions where it is mainly in charged form, the rates of hydrolysis would be directly proportional to the hydroxyl ion concentration. I n the intermediate region of the p H scale, the overall rate would depend upon the two rate constants and the dissociation constant of the base. For convenience the hydrolysis of the free base and the charged ion will be designated Reaction 1 and Reactim 2, respectively. Using the derivatims of these authors, the overall rate of hydrolysis of hornatropine may be expressed as
d _In _a dt
= -~ (OH-) kb
+ (OH-)
[k,(OH-)
+ k2kt,]
(Eq. 1)
where a = total drug concentratim, kb = dissociation constant of the base, kl = specific rate constant for Reaction 1 involving free base, and kl = specific rate constant for Resctim 2 involving the acid form. The right hand side of the above equation is constant for any given hydroxyl ion concentration. Thus, if Reactions 1 and 2 are mainly responsible for the hydrolysis of hcmatropine, the overall rate o f hydrolysis would be first order with rzspect to the ester concentration. A t any constant hydroxyl ion concentration, the half-life of the ester would then be expressed as
Expressed logarithmically, Eq. 2 becomes
879
A t relatively high hydroxyl ion concentration where (OH-) >>kb, Eq. 3 reduces to --log
tl/p
=
log (OH-)
4-log k1/0.693
(Eq. 4)
and a t low hydroxyl ion concentration where (OH-) kb, the equation reduces to
<<
-log
tl/n
= log (OH-)
+ log k2/0.693
(Eq. 5)
Thus at both ends of the pH scale, the rate of hydrolysis would be directly proportional to the hydroxyl ion concentration whereas there would be a region of nonproportionality in the interniediate range of p H scale. EXPERIMENTAL Spectrophotometric Analysis.-The method of analysis in this study consisted of making a series of ultraviolet absorption measurements on homatropine solution undergoing hydrolysis a t constant temperatures. Homatropine hydrobromide U. S. P . recrystallized from aqueous ethanol was used. Mandelic acid, the hydrolytic product of homatropine, absorbs practically a t the same wavelength as homatropine. Separation of mandelic acid was therefore necessary before scanning the solutions for residual ester determination. Absorbance measurements were made on a Beckman Model DU Spectrophotometer a t wavelengths of 257 and 263 mp. The ratio of the absorbdnces a t two wavelengths being constant over a wide range of concentration, the absorbances a t both the wavelengths were recorded for the first few experiments to serve a s a check on the analysis; for the rest of the experiments, absorbance was measured at wavelength 257 mp only. Buffer System.-To maintain hydroxyl ion concentration essentially constant, buffers or excess strong bases were used. When barium hydroxide solutions were used for low pOH values, the largest drop in hydroxyl ion concentration observed during the hydrolysis was 0.2 pOH unit. Half-life determinations were, however, made by analysis of the initial phase of the reaction during which there was no appreciable change in pH. Buffers used in the study are listed in Table I. The pH of each buffer was determined a t the temperature of the run. Standardized barium hydroxide solutions were used for pOH 2.2 and less.
TABLE I.-LIsT pH Range
OF
BUFFERS
Composition
i .18-i.94 M/15 KH2PO4-Na.,WPO4 8.23-8.84 M/10 HJBOl M/10 KCI 9.00-10.5 M/5 NHdOH HCI
++
+ SaOH
Experimental Procedure.-The following procedure with only minor modifications was employed in the present investigation : Buffer solution was placed in a thermostatically controlled bath set to the desired temperature and allowed to reach the temperature of the bath. Two ml. of standard solution containing sufficient homatropine hydrobromide was then added to give in the reaction mixture a concentration of approximately 1 mg./ml. After about three minutes when equilibrium was re-established, the first 10-ml. aliquot was with-
880
JOURNAL OF
THE
AMERICAN PHARMACEUTICAL ASSOCIATION
drawn, designated as the "0" minute sample and analyzed for residual cstcr. Samples, exactly 10 ml. in volume, were withdrawn a t suitable time intervals and residual ester was determined. The separation of mandelic acid from residual ester was done by a modification of the U. S. P. method for purification of alkaloids (7). The sample for analysis was made alkaline with ammonium hydroxide; the strength of ammonium hydroxide solution for each buffer being adjusted to extract the ester completely without appreciably hydrolyzing it. The base form of homatropine was immediately extracted with successive lo-, lo-, and 5-ml. portions of chloroform. From the combined chloroform extracts, the acid form of homatropine was extracted with exactly 10 ml. of 0.1 N HCI. The progress of the reaction was then calculated from the ultraviolet absorption of the acid extract. RESULTS AND DISCUSSION constant Influence of Ester Concentration.-At temperature and hydroxyl ion concentration, the observed rate of hydrolysis of hornatropine is first order with respect to the ester concentration. This is illustrated in Figs. 1, 2, and 3 representing, respectively, neutral, slightly alkaline, and alkaline reaction conditions. These results are in accord with Eq. 1. Influence of Hydroxyl Ion Concentration.-The effect of hydroxyl ion concentration on the rate of hydrolysis (expressed in terms of the half-life in minutes of the ester) is illustrated in Fig. 4. All the experimental points were determined a t or corrected to 30". The smooth curve in the intermediate pH region represents Eq. 3 where kl = 3.095 L. mole-' min.-', k2 = 123.2 L. mole-' min.-', and kb = 7.6 X The values for R1 and kf were obtained by substituting in Eqs. 4 and 5, respectively, the extreme experimental values at high and low pH in the curve. The kb was calculated from the pH of the half-neutralized solution of homatropine a t 30". Temperature Dependency.-Figure 4 indicates that above p H 11.75, the mechanism of hydrolysis is that of Reaction 1, or its equivalent; below pH 8.5 the mechanism is that of Reaction 2, or itsequivalent.
yo], S L y I I , KO, 13
20 40 60 TIME I N MINUTES
80
Fig. 2.-Hydrolysis of homatropine in solution a t pH 10.23 and at 30". I
zk
1.1)
LT:
r:
1.8
0
rl
1.7
8 1f i 2-1 TIME I N MIhWTES
32
40
Fig. 3.-Hydrolysis of homatropine in 0.01 iV Ba(0H)z and at 20'. Half-lives were determined in 0.01 N Ba(OH)2 solution and a phosphate buffer of pH 7.94 a t four different temperatures to study the temperature dependency of Reaction 1 and Reaction 2, respectively. The 1,)garithm of half-life periods in minutes is plotted against reciprocal of absolute temperature in Figs. 5 and 6. The straight line relationship indicates that the mechanisms responsible for the ester hydrolysis are not altered with changes in temperature. The activation energy was calculated by setting the slope equal to E/2.30R and the frequency factor(S) in the Arrhenius equation was calculated using k = S e c E I R T . The constants of the logarithmic form o f the Arrhenius equation are given in Table 11.
TARLEII.-TABULATIONOF CONSTAYTS OF THE LOGARITHMIC i \ R R € f E N I U S EQUATION"
10 20 T I M E I N HOURS
30
Fig. 1.-Hydrolysis of homatropine in phosphate buffer at pH 7.94 and a t 35.3".
Reacting Species
E . kcal / mole
Log
Homatropine (free base) Homatropine (charged ion)
12 3 11 46
9 4 10 5
+
s
" Loe S = loe k E / 2 303 R T where k IS ILI L mole I min. - 1 . b After subtracting 12 kcal./mole. the approximate heal of ionization of H20.
SCIENTIFIC EDITION
Jhxniher 19S.S
881
The simplified formula for the hydroxyl ion catalyzed hydrolysis of homatropine a t 30' may bc obtained by substituting the values a t 30" for k,, k 2 , and k b in Eq. 3 -log
tl/p
=
log (OH-) log [7.6 X lW5
+ (OH-)] +
In Tables 111, IV, and V, half-lives of homatropine solution determined experimentally are compared with those predicted by the above equation. Using activation energies, the rate constants a t any other temperature can be calculated. This, then establishes a basis for predicting the half-life period of homatropine a t various hydroxyl ion concentrations and temperatures. A n appreciable difference in rate constants is observed when the values for homatropine are compared with those of atropine. Ingold (5) has shown that monohydroxy ethyl acetate hydrolyzes ten times faster than ethyl acetate. The difference in rates has been attributed by this author clearly to the polar effect, whatever the stearic effects may be.
TABLEIII.-EXPERIMENTAL HALF-LIVES OF HOMAHIGHpH COMPARED WITH PREDICTED VALUES
TROPINE AT
Ternperature. O C .
POH
20
1 39
2.5
~.
30
35
1.65 2.00 2.00 2.00 2.20 2.00
--Half-Life, Lixperimental
Min.-------.
10.6 18.7 39.0 25.8 20.3 24.3 13.9
10.3 17.8 35.0 24.4 17.3 22.8 12.4
BY
Equation
TABLE IV.-EXPERIMENTAL HALF-LIVES OF HOMATROPINE AT NEUTRALpH COMPARED W I T H PREDICTED VALUES Temperature, O C .
30.0
35 3 40 8 46 9 a
PH"
-Half-l.ife. Bxperimental
7.18 7.38 7.54 7.94 7 94 7 94 7 94
332 . 0 221. B 166.0 63.1 31 4 1; n 7.9
Hours-
BY
Equation
349.5 221.0 153.2 61.7 32 3 16 6 8 1
2
3 -LOG
4 (OH-)
5
6
Fig. 4.-A curve relating the half-life of homatropine with the hydroxyl ion content of the reaction mixture a t 30". .-Experimental points determined a t 20" and corrected to 30".
1 .F,
1.5 h
'A
W b
s
5 1.4 -r >
s
1.3
1.2
pKw of water at 30° is 13.75.
3.2
3.3 1 x 100
3.1
TABLE V.-EXPERIMENTAL HALF-LIVES OF HOMAT TROPINE AT MODERATE pH A N D AT 30" COMPARED Fig. 5.-Temperature dependence of Reaction 1. WITH PREDICTED VALUES 1 Logarithm of half-life in minutes plotted against the reciprocal of absolute temperature of the run. Half-Life, Min.-PH
8.23 8.60 8.84 9-00 9.34 9.88 10.23 10.50
--
Experimental
By Equation
1794.0 871.5 483.5 372.0 202 0 113 0 84 7 61.5
1935.0 866.8 529.3 388.1 215 9 110 9 84 3 66.8
The study of the kinetics of the hydrolysis of atropine has been reported by Zvirblis, et al. (8). In the present investigation with homatropine, it is found that homatropine free base degrades a t a rate approximately five times that of atropine free base and the charged form of homatropine degrades a t a rate eight times that of the charged atropine. The inductive effect of the a-hydroxyl group in
882
4.0
JOURNAL OF THE
AMERICANPHARMACEUTICAL ASSOCIATION \‘d.XLVII. No. 12
i homatropine thus enhances the positive character of carbonyl carbon atom and accelerates the rate determining nucleophilic attack by hydroxide ion. This is in agreement with the scheme postulated by Ingold and Ingold (9) and by Waters (10) to represent the hydrolysis of ester by alkali. REFERENCES (1) Pittenger, P. S.. and Krantz, J . C . , Jr., THISJOURNAL, 17,1081(1928). (2) Schou. S. A., and Bjerregaard, P. B., D a m k Tidsskr. F a r m . , 6 , 185(1932). (3) Blok. C. J . , Pharm. J . . 155, 282(1915). (1) Hind, H . W., and Goyan, F. A f . , THISJOURNAL, 36, :33(1947). ( 5 ) Ingold. C. K., “Structure and Mechanism in Organic Chemistry,” Cornell University Press, Ithaca, N . Y., 1953, p.
,
?.,
Fig. 6.-Temperdture dependency of reaction 2. Logarithm of half-life in minutes plotted against the reciprocal of absolute temperature of the run.
(f;) Higuchi. T., Havinga, A., and Busse, L. W., THIS JOURNAL, 39,405(1950). ( 7 ) “United States Pharmacopeia,” Fourteenth Revision, Mack Publishing Co.. Easton, Pa., 1950, p. 681. (8) Zvirblis, P., Socholitsky, I.. and Kondritzer, A. A. THIS JOURNAL, 55, 450(1956). (9) Ingold, E . H . , and Ingold, C. K . , J . Chem. SOL., 756( 1932). (lo), Waters, W. A., “Physical Aspects of Organic Chemistry, 4th ed., D. Van Nostrand Co. Inc., New York, 1950, p. 332.
Glutarimides V* Synthesis of 2-Allyl-2-phenylglutarimide By MATHIAS P. MERTES, Jr. and CHARLES 0. WILSON Evidence is presented of lactone formation for 2-allyl-2-phenylglutaric acid. A synthesis for 2-allyl-2-phenylglutarimide is described and six compounds, which have not been found in the literature surveyed, are reported. The new compounds include 2-allyl-2-phenylglutaronitrile, 2-allyl-2-phenylglutaric acid, 2 - (2’,3’ - dibromopropyl) - 2 - phenylglutaronitrile, methyl 4-phenyl-4-cyano-6-heptenoate, 2-allyl-2-phenylglutaramic acid, and 2-allyl-2-phenylglutarimide. HERE HAVE BEEN four previous publications T(l-4) on glutarimide derivations. Their structural relationships to some useful pharmaceutical agents have been pointed out. The similarities in structure between compounds having analogous therapeutic activity have been the basis of investigation and synthesis of many therapeutic agents. In the barbiturate series there are six available compounds’ that are substituted in the five position with an unsaturated ally1 group. The hypnotic activity exhibited by noriden (glutethimide-Ciba), 2-ethyl-2-phenylglutarimide,has
* Received April 25, 1958, from the University of Texas, College of Pharmacy. Austin. Presented t o the Scientific Section, A. PH. A , , 1.0s Angeles meeting, April 1958. 1 These are: diallylbarbituric acid N . F. X.. aproharbital N . F. X . , kemithal. allylbarhituric acid N. P. X , Cyclopal, and secobarbital sodium U. S. P. XV.
established the glutarimide nucleus as having possible pharmacological activity, depending upon the substituents in the two position. Of the 2,2-disubstituted glutarimides, reported in the literature, there are no unsaturated alkyl derivatives. The preparation of 2,2-diallylglutarimide and 2-allyl-2-phenylglutarimide was undertaken. Several attempts to synthesize 2,2-diallylglutarimide were unsuccessful. By the method of Tagmann ( 5 ) 2-allyl-2-phenylglutaronitrile (I) was prepared. A ring closure method used by Tagmann ( 5 ) for preparing 2,2-disubstituted glutar imides by heating the substituted glutaronitrile I to 120° in sulfuric acid and acetic anhydride was not successful. The procedure of Benica and Wilson (1) was tried, starting with 2-allyl-2-phenylglutaric acid (II),which was prepared by hydrolysis of the dinitrile (I) in hydrochloric acid. Formation of substituted glutaric anhydrides usually proceeds with excellent yields as a result of refluxing with acetic anhydride. In the case of 2-allyl2-phenylglutaric acid (11) the starting material was recovered. The abnormal unreactivity of the 2-allyl-2phenylglutaric acid (11) was considered due to
878
from the multiple dose study B than an identical single dose of sustained release powder PD. The results of further studies with other compounds relating in vitro release and blood concentration will be reported. SUMMARY
1. Oral administration of a single 4-Gm. dose of SMTD resulted in rapid absorption of the drug with peak concentrations occurring in approximately one hour. 2. The rate of disappearance of the drug from blood could be approximated as first order during the early time intervals, permitting an estimation of the standard performance indexes k b and ( I ! ’ / & , . 3. For these subjects the average k b equaled 0 33, and ( t 1 / 8 ) b equaled two hours, 6 minutes for free SMTD. 4. When adult subjects received 1 Gm. of SMTD followed by 0.35 Gm. hourly for eight hours, steady state blood concentrations of free drug were maintained near 2-4 mg. yo during the one to eight-hour time interval. Reasonably uniurinary excretion Of from such a dosage regimen during the interval of administration. These data illustrate the use of kb in estimating the rate a t which drug must enter the blood in order to maintain fairly constant concentrations.
VijI,
S L V I I , No. 12
5 . The duration of absorption was approxi. mated in subjects receiving single doses of
‘
sustained suspensions to Gm. of SMTD. In subjects receiving powder PD peak blood concentrations occurred between two and four hours, and absorption was considered essentially complete after about four hours. Blood concentrations in the majority of the subjects receiving sustained release powder SD were more prolonged and the duration of absorption appeared to extend over most of the twelve-hour interval studied. 6. The results indicate a possible relationship between the rate of in vitro release and the duration of blood concentrations in human subjects, and the possible value of utilizing these relationships in formulation of some oral sustained release dosage forms. REFERENCES
(1) Foltz. E. L., Swintosky. J . v., and Robinson, M. J , FedernlroFr Proc.. 15 ( I ) , 422(195(i). ( 2 ) Swintosky, J. v., Robinson, M. J , Foitz. E. L . , and Free, S. M., THIS J O U R N A L , 46, 399(1957). (3) Swintosky V . , Robinson, M . J., and Foltz, E. L., jbi;ij v,, Foltz, E , L,, Bondi, A . , Jr., and
:$:$&:$;:
R o & ; ” ~ ; & ~ ~ ~ +; ,4 $ , , 4 ~ o ~ ~ ? , ( ~ ~Kobinson. ~ ~ , , M, J., ;bid., 7W19583. ( G ) Bratton. A. C . , and Marshall, E. K . , J r . , J . B i d . Chunr., 128, 537(1939). (7) Frisk, A. R . . Actn M e d . S c a n d . S i r p p l C ‘ X L I I , Jan., 19-13.
The Kinetics of the Hydrolysis of Homatropine* By J. L. PATELt and A. P. LEMBERGER An investigation of the chemical kinetics of the hydroxyl ion catalyzed hydrolysis of homatropine has shown that the mechanism of the hydrolytic deterioration consists of two reactions. The heat of activation for the two reactions involving the free and the acid form of the base as calculated from Arrhenius plots are 12.3 and 11.4 kilocalories per mole, respectively. It has been found that homatropine free base degrades at a rate approximately five times that of atropine free base, and the salt form of homatropine degrades at a rate approximately eight times that of the salt form of atropine.
of homatropine in solution is due mainly to the hydrolysis of the ester linkage yielding tropine and mandelic acid. Previous investigations (1-4) have shown that a t low pH homatropine is very stable even a t autoclave temperatures. A t higher pH it deteriorates and the rate of deterioration increases ETERIORATION
* Received April 25, 1958, from the School of Pharmacy, University of Wisconsin, hladison. t Present address: Alembic Chemical Works. Baroda. India. Supported in part by the Research Committee of the Graduate School from funds supplied by the Wisconsin Alumni Research Foundation. This paper is adapted from the prize-winning manuscript submitted by J. L. Patel in the Central Region, 1958. Lunsford Richardson Pharmacy Award competition. Presented to the Scientific Section, A.Pn.A.. Los Angeles meeting. April 1958.
uith temperature and hydroxyl ion concentration. The problem, however, has not been studied from the point of view of chemical kinetics. Many publications are found in the literature concerning the relative rates of second order alkaline hydrolysis of aliphatic carboxylic esters. Ingold ( 5 ) has shown the accelerating effect of C1, OH, COOMe, COMe, and CHzCOMe groups and the retarding effect of a negative ionic charge on the alkaline hydrolysis of aliphatic carboxylic esters. No studies, however, have been made to evaluate quantitatively the influence of substituent and environment upon the alkaline hydrolysis of highly substituted esters.
Decernber 1958
SCIENTIFIC EDITION
To evaluate quantitatively the contribution of polar and stearic factors and the positive ionic charge t o the rate of tropine ester hydrolysis, a study on t h e kinetics of the hydrolysis of homatropine, homatropine methyl bromide, and atropine methyl bromide was undertaken. This report presents a part of this investigation and deals specifically with t h e kinetics of the hydrolysis of homatropine. Another objective of this study was t o establish a basis for predicting the rate of hydrolysis of hornatropine i n solution at various hydroxyl ion concentrations and temperatures.
THEORETICAL CONSIDERATIONS The acid catalyzed hydrolysis of homatropine is slow. A study of 0.02 M homatropine in 0.05 M HC1 a t 25" showed no appearance of additional acid over a period of two days. An hydroxyl ion catalyzed reaction may then be the mechanism responsible for the hydrolysis of homatropine in solution. Homatropine can occur as free base or as the charged ion in solution. Two hydrolytic pathways are therefore possible. On the theoretical concepts employed by Higuchi, et al. (6). in their study on procaine, it is expected that in relatively high pH solutions where hornatropine is mainly present as free base and in low pH solutions where it is mainly in charged form, the rates of hydrolysis would be directly proportional to the hydroxyl ion concentration. I n the intermediate region of the p H scale, the overall rate would depend upon the two rate constants and the dissociation constant of the base. For convenience the hydrolysis of the free base and the charged ion will be designated Reaction 1 and Reactim 2, respectively. Using the derivatims of these authors, the overall rate of hydrolysis of hornatropine may be expressed as
d _In _a dt
= -~ (OH-) kb
+ (OH-)
[k,(OH-)
+ k2kt,]
(Eq. 1)
where a = total drug concentratim, kb = dissociation constant of the base, kl = specific rate constant for Reaction 1 involving free base, and kl = specific rate constant for Resctim 2 involving the acid form. The right hand side of the above equation is constant for any given hydroxyl ion concentration. Thus, if Reactions 1 and 2 are mainly responsible for the hydrolysis of hcmatropine, the overall rate o f hydrolysis would be first order with rzspect to the ester concentration. A t any constant hydroxyl ion concentration, the half-life of the ester would then be expressed as
Expressed logarithmically, Eq. 2 becomes
879
A t relatively high hydroxyl ion concentration where (OH-) >>kb, Eq. 3 reduces to --log
tl/p
=
log (OH-)
4-log k1/0.693
(Eq. 4)
and a t low hydroxyl ion concentration where (OH-) kb, the equation reduces to
<<
-log
tl/n
= log (OH-)
+ log k2/0.693
(Eq. 5)
Thus at both ends of the pH scale, the rate of hydrolysis would be directly proportional to the hydroxyl ion concentration whereas there would be a region of nonproportionality in the interniediate range of p H scale. EXPERIMENTAL Spectrophotometric Analysis.-The method of analysis in this study consisted of making a series of ultraviolet absorption measurements on homatropine solution undergoing hydrolysis a t constant temperatures. Homatropine hydrobromide U. S. P . recrystallized from aqueous ethanol was used. Mandelic acid, the hydrolytic product of homatropine, absorbs practically a t the same wavelength as homatropine. Separation of mandelic acid was therefore necessary before scanning the solutions for residual ester determination. Absorbance measurements were made on a Beckman Model DU Spectrophotometer a t wavelengths of 257 and 263 mp. The ratio of the absorbdnces a t two wavelengths being constant over a wide range of concentration, the absorbances a t both the wavelengths were recorded for the first few experiments to serve a s a check on the analysis; for the rest of the experiments, absorbance was measured at wavelength 257 mp only. Buffer System.-To maintain hydroxyl ion concentration essentially constant, buffers or excess strong bases were used. When barium hydroxide solutions were used for low pOH values, the largest drop in hydroxyl ion concentration observed during the hydrolysis was 0.2 pOH unit. Half-life determinations were, however, made by analysis of the initial phase of the reaction during which there was no appreciable change in pH. Buffers used in the study are listed in Table I. The pH of each buffer was determined a t the temperature of the run. Standardized barium hydroxide solutions were used for pOH 2.2 and less.
TABLE I.-LIsT pH Range
OF
BUFFERS
Composition
i .18-i.94 M/15 KH2PO4-Na.,WPO4 8.23-8.84 M/10 HJBOl M/10 KCI 9.00-10.5 M/5 NHdOH HCI
++
+ SaOH
Experimental Procedure.-The following procedure with only minor modifications was employed in the present investigation : Buffer solution was placed in a thermostatically controlled bath set to the desired temperature and allowed to reach the temperature of the bath. Two ml. of standard solution containing sufficient homatropine hydrobromide was then added to give in the reaction mixture a concentration of approximately 1 mg./ml. After about three minutes when equilibrium was re-established, the first 10-ml. aliquot was with-
880
JOURNAL OF
THE
AMERICAN PHARMACEUTICAL ASSOCIATION
drawn, designated as the "0" minute sample and analyzed for residual cstcr. Samples, exactly 10 ml. in volume, were withdrawn a t suitable time intervals and residual ester was determined. The separation of mandelic acid from residual ester was done by a modification of the U. S. P. method for purification of alkaloids (7). The sample for analysis was made alkaline with ammonium hydroxide; the strength of ammonium hydroxide solution for each buffer being adjusted to extract the ester completely without appreciably hydrolyzing it. The base form of homatropine was immediately extracted with successive lo-, lo-, and 5-ml. portions of chloroform. From the combined chloroform extracts, the acid form of homatropine was extracted with exactly 10 ml. of 0.1 N HCI. The progress of the reaction was then calculated from the ultraviolet absorption of the acid extract. RESULTS AND DISCUSSION constant Influence of Ester Concentration.-At temperature and hydroxyl ion concentration, the observed rate of hydrolysis of hornatropine is first order with respect to the ester concentration. This is illustrated in Figs. 1, 2, and 3 representing, respectively, neutral, slightly alkaline, and alkaline reaction conditions. These results are in accord with Eq. 1. Influence of Hydroxyl Ion Concentration.-The effect of hydroxyl ion concentration on the rate of hydrolysis (expressed in terms of the half-life in minutes of the ester) is illustrated in Fig. 4. All the experimental points were determined a t or corrected to 30". The smooth curve in the intermediate pH region represents Eq. 3 where kl = 3.095 L. mole-' min.-', k2 = 123.2 L. mole-' min.-', and kb = 7.6 X The values for R1 and kf were obtained by substituting in Eqs. 4 and 5, respectively, the extreme experimental values at high and low pH in the curve. The kb was calculated from the pH of the half-neutralized solution of homatropine a t 30". Temperature Dependency.-Figure 4 indicates that above p H 11.75, the mechanism of hydrolysis is that of Reaction 1, or its equivalent; below pH 8.5 the mechanism is that of Reaction 2, or itsequivalent.
yo], S L y I I , KO, 13
20 40 60 TIME I N MINUTES
80
Fig. 2.-Hydrolysis of homatropine in solution a t pH 10.23 and at 30". I
zk
1.1)
LT:
r:
1.8
0
rl
1.7
8 1f i 2-1 TIME I N MIhWTES
32
40
Fig. 3.-Hydrolysis of homatropine in 0.01 iV Ba(0H)z and at 20'. Half-lives were determined in 0.01 N Ba(OH)2 solution and a phosphate buffer of pH 7.94 a t four different temperatures to study the temperature dependency of Reaction 1 and Reaction 2, respectively. The 1,)garithm of half-life periods in minutes is plotted against reciprocal of absolute temperature in Figs. 5 and 6. The straight line relationship indicates that the mechanisms responsible for the ester hydrolysis are not altered with changes in temperature. The activation energy was calculated by setting the slope equal to E/2.30R and the frequency factor(S) in the Arrhenius equation was calculated using k = S e c E I R T . The constants of the logarithmic form o f the Arrhenius equation are given in Table 11.
TARLEII.-TABULATIONOF CONSTAYTS OF THE LOGARITHMIC i \ R R € f E N I U S EQUATION"
10 20 T I M E I N HOURS
30
Fig. 1.-Hydrolysis of homatropine in phosphate buffer at pH 7.94 and a t 35.3".
Reacting Species
E . kcal / mole
Log
Homatropine (free base) Homatropine (charged ion)
12 3 11 46
9 4 10 5
+
s
" Loe S = loe k E / 2 303 R T where k IS ILI L mole I min. - 1 . b After subtracting 12 kcal./mole. the approximate heal of ionization of H20.
SCIENTIFIC EDITION
Jhxniher 19S.S
881
The simplified formula for the hydroxyl ion catalyzed hydrolysis of homatropine a t 30' may bc obtained by substituting the values a t 30" for k,, k 2 , and k b in Eq. 3 -log
tl/p
=
log (OH-) log [7.6 X lW5
+ (OH-)] +
In Tables 111, IV, and V, half-lives of homatropine solution determined experimentally are compared with those predicted by the above equation. Using activation energies, the rate constants a t any other temperature can be calculated. This, then establishes a basis for predicting the half-life period of homatropine a t various hydroxyl ion concentrations and temperatures. A n appreciable difference in rate constants is observed when the values for homatropine are compared with those of atropine. Ingold (5) has shown that monohydroxy ethyl acetate hydrolyzes ten times faster than ethyl acetate. The difference in rates has been attributed by this author clearly to the polar effect, whatever the stearic effects may be.
TABLEIII.-EXPERIMENTAL HALF-LIVES OF HOMAHIGHpH COMPARED WITH PREDICTED VALUES
TROPINE AT
Ternperature. O C .
POH
20
1 39
2.5
~.
30
35
1.65 2.00 2.00 2.00 2.20 2.00
--Half-Life, Lixperimental
Min.-------.
10.6 18.7 39.0 25.8 20.3 24.3 13.9
10.3 17.8 35.0 24.4 17.3 22.8 12.4
BY
Equation
TABLE IV.-EXPERIMENTAL HALF-LIVES OF HOMATROPINE AT NEUTRALpH COMPARED W I T H PREDICTED VALUES Temperature, O C .
30.0
35 3 40 8 46 9 a
PH"
-Half-l.ife. Bxperimental
7.18 7.38 7.54 7.94 7 94 7 94 7 94
332 . 0 221. B 166.0 63.1 31 4 1; n 7.9
Hours-
BY
Equation
349.5 221.0 153.2 61.7 32 3 16 6 8 1
2
3 -LOG
4 (OH-)
5
6
Fig. 4.-A curve relating the half-life of homatropine with the hydroxyl ion content of the reaction mixture a t 30". .-Experimental points determined a t 20" and corrected to 30".
1 .F,
1.5 h
'A
W b
s
5 1.4 -r >
s
1.3
1.2
pKw of water at 30° is 13.75.
3.2
3.3 1 x 100
3.1
TABLE V.-EXPERIMENTAL HALF-LIVES OF HOMAT TROPINE AT MODERATE pH A N D AT 30" COMPARED Fig. 5.-Temperature dependence of Reaction 1. WITH PREDICTED VALUES 1 Logarithm of half-life in minutes plotted against the reciprocal of absolute temperature of the run. Half-Life, Min.-PH
8.23 8.60 8.84 9-00 9.34 9.88 10.23 10.50
--
Experimental
By Equation
1794.0 871.5 483.5 372.0 202 0 113 0 84 7 61.5
1935.0 866.8 529.3 388.1 215 9 110 9 84 3 66.8
The study of the kinetics of the hydrolysis of atropine has been reported by Zvirblis, et al. (8). In the present investigation with homatropine, it is found that homatropine free base degrades a t a rate approximately five times that of atropine free base and the charged form of homatropine degrades a t a rate eight times that of the charged atropine. The inductive effect of the a-hydroxyl group in
882
4.0
JOURNAL OF THE
AMERICANPHARMACEUTICAL ASSOCIATION \‘d.XLVII. No. 12
i homatropine thus enhances the positive character of carbonyl carbon atom and accelerates the rate determining nucleophilic attack by hydroxide ion. This is in agreement with the scheme postulated by Ingold and Ingold (9) and by Waters (10) to represent the hydrolysis of ester by alkali. REFERENCES (1) Pittenger, P. S.. and Krantz, J . C . , Jr., THISJOURNAL, 17,1081(1928). (2) Schou. S. A., and Bjerregaard, P. B., D a m k Tidsskr. F a r m . , 6 , 185(1932). (3) Blok. C. J . , Pharm. J . . 155, 282(1915). (1) Hind, H . W., and Goyan, F. A f . , THISJOURNAL, 36, :33(1947). ( 5 ) Ingold. C. K., “Structure and Mechanism in Organic Chemistry,” Cornell University Press, Ithaca, N . Y., 1953, p.
,
?.,
Fig. 6.-Temperdture dependency of reaction 2. Logarithm of half-life in minutes plotted against the reciprocal of absolute temperature of the run.
(f;) Higuchi. T., Havinga, A., and Busse, L. W., THIS JOURNAL, 39,405(1950). ( 7 ) “United States Pharmacopeia,” Fourteenth Revision, Mack Publishing Co.. Easton, Pa., 1950, p. 681. (8) Zvirblis, P., Socholitsky, I.. and Kondritzer, A. A. THIS JOURNAL, 55, 450(1956). (9) Ingold, E . H . , and Ingold, C. K . , J . Chem. SOL., 756( 1932). (lo), Waters, W. A., “Physical Aspects of Organic Chemistry, 4th ed., D. Van Nostrand Co. Inc., New York, 1950, p. 332.
Glutarimides V* Synthesis of 2-Allyl-2-phenylglutarimide By MATHIAS P. MERTES, Jr. and CHARLES 0. WILSON Evidence is presented of lactone formation for 2-allyl-2-phenylglutaric acid. A synthesis for 2-allyl-2-phenylglutarimide is described and six compounds, which have not been found in the literature surveyed, are reported. The new compounds include 2-allyl-2-phenylglutaronitrile, 2-allyl-2-phenylglutaric acid, 2 - (2’,3’ - dibromopropyl) - 2 - phenylglutaronitrile, methyl 4-phenyl-4-cyano-6-heptenoate, 2-allyl-2-phenylglutaramic acid, and 2-allyl-2-phenylglutarimide. HERE HAVE BEEN four previous publications T(l-4) on glutarimide derivations. Their structural relationships to some useful pharmaceutical agents have been pointed out. The similarities in structure between compounds having analogous therapeutic activity have been the basis of investigation and synthesis of many therapeutic agents. In the barbiturate series there are six available compounds’ that are substituted in the five position with an unsaturated ally1 group. The hypnotic activity exhibited by noriden (glutethimide-Ciba), 2-ethyl-2-phenylglutarimide,has
* Received April 25, 1958, from the University of Texas, College of Pharmacy. Austin. Presented t o the Scientific Section, A. PH. A , , 1.0s Angeles meeting, April 1958. 1 These are: diallylbarbituric acid N . F. X.. aproharbital N . F. X . , kemithal. allylbarhituric acid N. P. X , Cyclopal, and secobarbital sodium U. S. P. XV.
established the glutarimide nucleus as having possible pharmacological activity, depending upon the substituents in the two position. Of the 2,2-disubstituted glutarimides, reported in the literature, there are no unsaturated alkyl derivatives. The preparation of 2,2-diallylglutarimide and 2-allyl-2-phenylglutarimide was undertaken. Several attempts to synthesize 2,2-diallylglutarimide were unsuccessful. By the method of Tagmann ( 5 ) 2-allyl-2-phenylglutaronitrile (I) was prepared. A ring closure method used by Tagmann ( 5 ) for preparing 2,2-disubstituted glutar imides by heating the substituted glutaronitrile I to 120° in sulfuric acid and acetic anhydride was not successful. The procedure of Benica and Wilson (1) was tried, starting with 2-allyl-2-phenylglutaric acid (II),which was prepared by hydrolysis of the dinitrile (I) in hydrochloric acid. Formation of substituted glutaric anhydrides usually proceeds with excellent yields as a result of refluxing with acetic anhydride. In the case of 2-allyl2-phenylglutaric acid (11) the starting material was recovered. The abnormal unreactivity of the 2-allyl-2phenylglutaric acid (11) was considered due to