Journal of Clinical Pharmacy (1978) 3,l-6.
`
`STABILITY OF COCAINE IN AQUEOUS SOLUTION
`
`John B. Murray* and Hasan I. AlShora
`*Heriot- Watt University, Pharmacy Department, 79 Grassmarket. Edinburgh EHI 2HJ
`
`S U M M A R Y
`The effect of pH on the rate of hydrolysis of cocaine was investigated. Ionic
`strength had a negligible effect while the concentrations of the various buffers
`had a significant effect on the rate of hydrolysis. The pH rate profile in buffer
`showed a minimum at 1.95. The energies of activation were determined to be
`95.9, 107.3 and 97.9 kJ mol-' respectively at pH 6.8,51) and 2.2. The stability
`on storage of sterilized solutions of cocaine prepared by various methods and
`formulae was compared with the stability predicted from the results of the
`accelerated tests.
`
`Cocaine is a naturally occurring alkaloid obtained from the dried leaves of Eryfhroxylum
`coca or E. fruxillense. In addition, the leaves contain benzoylecgonine, ecgonine and a
`variety of other alkaloids (Fig. 1).
`A number of studies on the degradation of cocaine have been reported since the late
`19th century, but none of these authors isolated cocaine from benzoylecgonine and ecgonine
`before measuring the amount present and the degradation products were included in and
`responded to the assay.
`It is used as a local anaesthetic in eye, nose and throat surgery and in euphoriant elixirs
`which are given to patients with terminal carcinoma.
`
`MATERIALS
`Cocaine hydrochloride was obtained from Macarthys Ltd., Edinburgh, and Macfarlan Smith
`Ltd., Edinburgh. Ecgonine was donated by May and Baker Co. Ltd, Dagenham, England.
`Water was prepared from an all-glass still. All other chemicals were reagent grade and those
`used as buffers were Analar.
`
`ASSAY PROCEDURE
`A sample (0.25 ml containing 2.5 mg of cocaine hydrochloride) was mixed with 10 ml of
`5% sodium bicarbonate and extracted with ether (10, 10 and 15 ml). The combined extract
`was dried over sodium sulphate anhydrous and then extracted with 0.01 N HCl(lO0 and 50
`ml). The combined aqueous extracts 'were diluted to 250 ml with 0.01 N HCl and the
`0308-6593/78/0300-0001$02.00 01978 Blackwell Scientific Publications
`
`1
`
`MYLAN ET AL. - EXHIBIT 1016
`
`

`
`2
`
`John B. Murray and Hasan I. Al-Shora
`
`CH3
`
`COCAINE
`
`RI
`-COOCHS
`
`BENZOYLECGONINE
`-COOH
`
`R2
`0
`
`-
`O
`
`C
`
`~
`
`0
`
`o
`C
`-
`
`~
`
`-
`
`-
`
`
`
`
`
`ECGONINE
`-OH
`-COOH
`Fig. 1. Structure of main alkaloids of Erythrotylum spp.
`
`extinction measured at 233 nm against 0.01 N HC1. Mean absorbance of ten experiments
`0.381. Standard deviation (s.d.) 0.0022 and coefficient of variation (C.V.) 0.56%.
`To ascertain that benzoylecgonine and ecgonine did not interfere with the assay of cocaine,
`cocaine 1% was measured in presence of 0.5% benzoylecgonine and 0.5% ecgonine. The
`mean absorbance = 0.382, s.d. = 0.003, and C.V. = 0.79%. Beer’s Law was found to be
`obeyed.
`
`KINETIC PROCEDURE
`To determine the effects of pH on the rate of degradation of cocaine a solution of 1%
`cocaine hydrochloride in the appropriate buffer solution was prepared and immediately
`distributed into 5-ml glass ampoules and then immersed in a water bath at constant tempera-
`ture. Two minutes were permitted for the temperature to equilibrate and then a sample
`was taken and cooled in ice and two 2-ml portions were assayed by the procedure quoted
`previously. Further samples of the solutions were taken at six appropriate times and assayed.
`
`RESULTS AND DISCUSSION
`Under the above conditions cocaine degraded to benzoylecgonine, no ecgonine was found
`on the basis of thin-layer chromatography evidence.
`
`Order of reaction and observed rate constant
`At constant pH and temperature the degradation of cocaine was found to observe pseudo
`first-order kinetics. Fig. 2 shows the results of several experimental runs at 80°C and at
`various pHs. The observed rate constants kobs between pH 1 and pH 8.8 are shown (Table 1).
`A minimum rate was observed at pH 1.95. The observed rate was actually a summation of
`a series of catalytic reaction rates induced by the buffer species, hydrogen and hydroxide
`ions and water molecules.
`General acid-base catalysis was observed and the rates of degradation of cocaine in citric
`acid-potassium citrate buffer solutions (Table 2) increased with increased concentration of
`buffer between pH 2.2 and 4.8.
`Both acidic and alkaline hydrolysis is possible for cocaine. In solution it can exist as the
`
`

`
`Cocaine stability
`
`3
`
`2.0 cl
`
`1.4
`
`I
`5
`
`I
`10
`
`\
`
`I
`I
`20
`15
`Time (min )
`
`I
`25
`
`I
`30
`
`Fig. 2. Plot of log percentage residual concentration against time for solutions of
`cocaine hydrochloride at 8OoC and pH 7.65 (A), pH 7.00 (0) and pH 6-48 (0).
`
`Table 1. Stability at 8OoC of a 0.1% solution of cocaine hydrochloride
`from pH 8.8 to pH 1.0 in single strength buffer*
`
`PH
`
`8.80
`8.45
`8.00
`7.65
`7.00
`6.48
`6.00
`5.60
`5.20
`4.95
`4.55
`4.25
`3.60
`3.18
`2.6 1
`1.95
`1.30
`1 .no
`
`Rate constant
`k (s-’)
`
`0.4604 X
`0.3985 X
`0.2480 X
`0.1691 X lo-’
`0.6747 X
`0.3160 X
`0.1294 X
`0.3854 X
`0.1662 X
`0.9675 X lo-’
`0.4898 X lo-’
`0.3180X
`0.1695 X
`0.1035 X
`0.7274 X
`0.4757 X
`0.1798X 10.‘
`0.4475 X
`
`Standard error
`of rate constant (s-’)
`
`0.1679X
`0.1494 X
`0.5804 X
`0.3257 X
`0.8506 X lo-’
`0.6148X lo-’
`0.8426 X
`0.4000 X
`0.2333 X
`0.3634 X
`0.8716 X
`0.6227 X
`0.2843 X lo-’
`0.4558 X lo-’
`0 . 1 1 2 2 ~ 10-7
`0.1285 X
`0.1452 X
`0.2636 X lo-’
`
`*pH 1-1.85 Clark and Lubs potassium chloride-hydrochloric acid
`buffer. pH 2.2-8.0 McIlwaine’s citric acid-phosphate buffer. pH 8.0
`Bates and Bower borax-hydrochloric acid buffer.
`
`

`
`Q
`3
`a
`
`ta
`3 x
`3
`
`P
`
`-5.278
`
`-5.688
`
`-6.125
`
`-6.346
`
`5-27 -0.02
`+0.001
`2.05 4.14
`+0.21
`Oq8' -0.08
`
`+0.04
`
`0.45 -0.03
`+0.02
`
`*The rate at buffer-free condition was obtained by extrapolation.
`
`10.60 f 0.63
`
`4.17 f 0.09
`
`1.84 f 0.06
`
`1.05 f 0.01
`
`9.46 f 0.21
`
`3.93 f 0.1
`
`1.59 f 0.06
`
`7.94 f 0.32
`
`3.21 f 0.05
`
`1.25 f 0.03
`
`6.63 f 0.25
`
`2.58 f 0.1
`
`0.881 f 0.03
`
`0.918 f 0.02
`
`0.729 f 0.01
`
`0.569 f 0.01
`
`5.94 f 0.13
`
`2.21 i 0.04
`
`0.971 f 0.05
`
`0.494 f 0.02
`
`4.8
`
`4.2
`
`3.2
`
`2.2
`
`log KO*
`
`KO+
`
`0.4
`
`0.3
`
`0.2
`
`0.1
`
`0.05
`
`PH
`
`Buffer concentration
`
`Table 2. Effects of the molar concentration of citric acid-potassium citrate buffer solution on the stability of a 1.0% solution of cocaine hydrochloride at 80°C
`
`s-' f confidence limits at total buffer concentration mol/l
`
`and various pH values. Observed rate constant x
`
`

`
`Cocaine stability
`5
`free base and as the monoprotonated form, the concentration of these two species being
`dependent upon the pH of the solution and the ionization constant of cocaine.
`
`The effect of ionic strength
`The rate constants at pH 1.15,5.0 and 6.0 were compared with rate constants at the same
`pHs obtained with 1.169% NaCl present and the differences were found to be neglible.
`
`Apparent heat of activation
`The temperature dependence of the hydrolytic rate constant of cocaine in aqueous solu-
`tion was shown by the results obtained from pH 6.8, 5.0 and 2.2 at a range of temperatures;
`the Arrhenius type plots are shown in Fig. 3.
`
`'
`
`-81
`2-7 2.8
`
`1
`'
` 1
`I
`2.9 3.0 3.1 3.2
`fx10-3
`(K-') for a 1.0% solution of
`Fig. 3. Plot of the log rate constant k(s-') against
`cocaine hydrochloride at pH 6.8 (A), pH 5.0 (a) and pH 2.2 (0).
`
`'
`
`I
`3.3 3.4
`
`The calculated energies of activation are 95.9 f 3.7 kJ mol-' at pH 6 8 , 107.3 f 6 0
`kJ mol-' at pH 5.00 and 97.9 f 9.3 kJ mol-' at pH 2.2. These values are in the same range
`as those quoted for other ester hydrolyses although direct comparisons are not possible
`unless the determinations are made at the same pH values.
`
`PHARMACEUTICAL APPLICATIONS
`The main application of the work has been the determination of the stability of Cocaine
`Eye Drops BPC and a comparison with buffered cocaine solutions. The results are shown in
`Tables 3 and 4.
`Cocaine Eye Drops BPC are unbuffered sterile solutions, the stability of which after
`autoclaving or steaming in 5-ml glass ampoules is shown in Table 3.
`At pH 6.8 in Table 4 and 25°C the predicted tl% was 24.5 h + 1 .l, -1.06 (P = 0.95)
`and the experimental f was in the range 23.4-22.7 h. However, at 5°C the predicted and
`experimental tlosb were not in close agreement, the difference being possibly due to the
`difficulty in control of the storage temperature at 5°C. At pH 5.0 good agreement was
`
`

`
`John B. Murray and Hasan I. Al-Shora
`Table 3. Stability of Cocaine Eye Drops BPC
`
`Initial
`cocaine HCI
`concentration
`
`5%
`5% (*)
`
`Initial
`PH
`
`5.8
`5.8
`
`Cocaine HCI concentration as % of initial concentration
`After 3
`After 6
`months
`months
`
`5°C
`
`25°C
`
`5°C
`
`25°C
`
`pH after
`sterilization
`
`After
`sterilization
`
`3.1
`-
`
`97
`-
`
`96
`97
`
`90
`93
`
`89
`91
`
`80
`89
`
`* This solution was not sterilized: the pH after 6 months was 3.0. The drop in the pH of the solution can
`be attributed to the benzoylecgonine produced during the sterilization procedure, each individual solution
`shows a constant pH during storage thereafter. The results in Table 4 are for 1% cocaine hydrochloride
`solutions, buffered with HCl or citrate-phosphate buffers, and sterilized by filtration. Similar results were
`obtained with 0.25,0.5, 2 and 4% solutions.
`
`Table 4. Stability of cocaine solution buffered at pH 6.8,5.0 and 2.2, sterilized by filtration
`
`Cocaine HCl concentration as % of initial concentration
`After 6 months
`After 3 months
`After 1 week
`
`5°C
`
`88
`-
`-
`
`25°C
`
`46
`-
`-
`
`5°C
`
`-
`99
`100
`
`25°C
`
`5°C
`
`25°C
`
`-
`90
`98
`
`-
`96
`99
`
`-
`80
`95
`
`Initial
`
`6.8
`5 .o
`2.2
`
`found at 25°C between the experimental and predicted rlm. The predicted rlW was 102.1
`+ 39.4/- 30.0 days and the found rlW was 87.0 f 1.9 days. Storage at 5°C for pH 5.0
`showed only a 4% loss after 6 months.
`Cocaine eye drops buffered at pH 2.2 were very stable losing only 5% after 6 months
`storage at 25°C. Although cocaine hydrochloride solutions buffered at pH 5 and at pH 2.2
`exhibit good stability they may cause irritation before the eye can adjust the pH to 7.4.
`As cocaine is absorbed and acts as the free base, and with a pKa of 8.6 the quicker the pH
`increases the better for the anaesthetic properties of the solution.
`In conclusion we consider that the formulation and methods of preparation recom-
`mended by the BPC 1973 are acceptable in respect of the chemical stability of the alkaloid:
`the assay outlined above should replace the BPC assay and the eye drops should be stored
`at 5°C with a time limit of 6 months.