Photochemical decomposition of phenazone derivatives. Part 7: Mechanism of decomposition in aqueous solutions

Kinetic studies, supplemented by the isolation and identification of the products of decomposition, have proved the process of photochemical decomposition of phenazone derivatives to be a complex reaction, involving several successive parallel reactions, one of which is predominant depending on the concentration of the solution and the atmosphere above it. In an oxygen-free atmosphere and at low concentrations (10(-4) mol/dm3) decomposition is almost wholly the result of second photolysis of zero order to aziridine derivative----aniline----isonitrile. At higher concentrations (10(-3)-10(-2) mol/dm3), the contribution from the reaction of water photoaddition to the double bond C3-C4 increases. Whereas at concentrations of order greater than or equal to 10(-2) mol/dm3, photoisomerization to imidazole derivatives is predominant. In air, one onserves additionally a second reaction of photooxidation to 4-hydroxy-phenazone----1-acetylo-1-methyl-2-phenyl-hydrazine, a reaction of photodemethylation at N2 with simultaneous oxidation to the 4-ketoderivative, and reactions of hydrolysis characteristic for the individual derivatives. The principal primary photolytic reaction for the group of compounds studied consists in cleavage of the N1-N2 bond of the pyrazoline ring.     

Effect of cyclodextrin complexation on aqueous solubility and photostability of phenothiazine

The effect of cyclodextrin (beta-CD, gamma-CD and substituted beta-CD derivatives) complexation on the solubility and photostability of phenothiazine (Ph) was compared. The phase solubility method was applied to calculate the stability constants of soluble 1:1 or 1:2 inclusion compounds formed between Ph and CDs. Photochemical decomposition in solution of phenothiazine alone and in the presence of beta-CD or beta-CD derivatives, was found to proceed according to the two stage first-order reaction and in the case of gamma-CD, in a single stage reaction. Formation of solid inclusion complexes of Ph with CDs was evaluated using IR, 13C NMR and DSC studies. The influence of the complexation technique in the solid state (kneading, heating and freeze-drying) on the solubility of Ph was compared. It was establish that the improvement in solubility and stability of Ph was dependent on the kind of CD. When the complexation proceeded in solution it was more effective.     

Studies on the photostability and in vitro phototoxicity of Labetalol

The purpose of this study was to obtain information on the photochemical and phototoxic properties of Labetalol, a beta-blocker drug. Preliminary information on the drug photoreactivity was achieved using a flow system with a photochemical reactor on-line with a diode array detection system. Photophysical and photochemical investigations on the drug were performed in aqueous solutions at different pH values using spectrophotometric and fluorimetric methods; the photodegradation quantum yield was found to be 2.7×10⁻³ at pH 5.8 and 1.5×10⁻² at pH 11.5. Forced photodegradation of labetalol solutions under exposure to UVA–UVB radiations (xenon arc lamp) was monitored by reversed-phase liquid chromatography. The main photodegradation products were isolated and characterized by NMR and mass spectrometry; labetalol was found to give 3-amino-1-phenylbutane and salicylamide-4-carboxaldehyde as the main photoproducts. Preliminary phototoxic testings on human keratinocyte cultures were performed evaluating the viability of the cells by the neutral-red uptake assay; mutagenic and photomutagenicity tests were also carried out based on Salmonella typhimurium strains. As a result, labetalol was found to be photolabile,mainly in alkaline medium, but evidences of significant phototoxic and photomutagenic effects by the drug were not observed.     

Determination of photostability and photodegradation products of indomethacin in aqueous media

Photochemical behaviour of indomethacin in aqueous media at 254 nm, 310 nm and sunlight was studied by HPLC. The drug exhibited a similar behaviour in all the irradiation experiments affording eight photoproducts that were separated and identified. The main photochemical routes are suggested to proceed via decarboxylation, followed by oxygenation to give an alcohol and an aldehyde and/or by solvent trapping to produce the alcohol. Photoinduced hydrolysis of CO-N bond and oxidative C2-C3 bond breakage also occur.     

Studies on the photostability and phototoxicity of aloe-emodin,emodin and rhein

Aloe-emodin (1), emodin (2) and rhein (3) were found to be photolabile by visible (390-500 nm) light under aerobic conditions. The drugs 1, 2 and 3 were phototoxic in vitro when examined by the photohemolysis test under both oxygen and argon atmospheres, although the photohemolysis rate was markedly lower under anaerobic conditions. The experiments were also carried out in the presence of butylated hydroxyanisole (BHA), reduced glutathione (GSH), sodium azide (NaN3) and superoxide dismutase (SOD). Based on the inhibition of this process on addition of BHA, GSH, SOD and NaN3, there would seem to be involvement of free radicals (type I mechanism) and singlet oxygen in the process (type II mechanism). The in vitro phototoxicity of this anthraquinone series was also verified in a lipid-photoperoxidation test with linoleic acid. In summary, this anthraquinone series is phototoxic in vitro. This behavior can be explained through the involvement of singlet oxygen and stable photoproducts.     

Stability of ertapenem in aqueous solutions

The kinetics of degradation of ertapenem was studied in aqueous solutions at 303, 313, 323 and 333 K and pH 0.42-12.5. Degradation was studied using two methods: HPLC (LiChrospher RP-18 column, 5 microm, 250 mm x 4 mm; mobile phase: methanol-phosphate buffer 25 mmol l(-1), pH 6.5 (15:85, v/v); flow rate--1.2 ml/min; detection UV--298 nm) and UV (294 nm). Specific acid-base catalysis involves: (a) hydrolysis of ertapenem, catalysed by hydrogen ions; (b) hydrolysis of ertapenem dianions catalysed by hydroxide ions; (c) spontaneous hydrolysis of zwitter ions and dianions of ertapenem under the influence of water. The thermodynamic parameters of these reactions--energy, enthalpy and entropy of activation were calculated. It was observed that buffer catalysis occurred in acetate, phosphate and borate buffers.     

Stability of galactose in aqueous solutions

The stability of 5%-30% w/v galactose in sterile water for injection and acetate and phosphate buffers was studied. The concentration of galactose was determined after each sample was diluted to a nominal concentration of 0.5% (w/v); for purposes of data analysis, the concentration as measured in the diluted sample was multiplied by a dilution factor to obtain the true concentration in the sample. The concentrations were determined from the regression line obtained by plotting the peak-height ratios (for various concentrations of galactose and the internal standard cellobiose) versus the galactose concentrations. Triplicate samples were quantitatively analyzed for galactose content by high-performance liquid chromatography. The stability of the samples was then studied in relation to buffer concentration; pH; storage at 25, 45, and 65 degrees C for six weeks, and autoclaving at 121 degrees C for 30 minutes. Galactose degradation increased in relation to its concentration, increasing temperature, and buffer concentration. Galactose solutions in water and phosphate incurred less than 5% degradation on autoclaving; however, the 30% solutions in acetate buffers lost up to 21% of initial content. Yellow discoloration of solutions was associated with autoclaving and prolonged exposure at 65 degrees C and appeared in some solutions that did not exceed the USP XXI limit of 5-hydroxymethylfurfural and related compounds in dextrose injection. The estimated room temperature shelf-life of galactose in sterile water for injection sterilized by 0.45-micron-porosity membrane filtration is four and one-half months. Solutions may also be sterilized by autoclaving at 121 degrees C for 30 minutes; galactose solutions containing pH buffers should not be sterilized by autoclaving.     

Viscosity effects on nebulisation of aqueous solutions

While the viscosity of sucrose and sodium citrate solutions increased with increasing percentage (w/v) of solute, the surface tension largely remained consistent. Water, sucrose (10–60% w/v) and sodium citrate (7–36% w/v) solutions were nebulised in two air-jet nebulisers (Pari LC; Medix A II) operated at 6 1 min⁻¹ and in an ultrasonic device (Medix Electronic) operated at mid-power setting. A wide variation in nebuliser size and output characteristics existed between the different commercially available models studied. Contrary to atomisation theories for air-jet nebulisation, droplet size was inversely related to solution viscosity (over 1–6 cP). Beyond this critical value droplet size increased as viscosity increased. This anomaly may be due to viscosity acting either directly or through liquid flow-rates. By contrast the ultrasonic device generated droplets of size proportional to viscosity but was unable to nebulise high viscosity fluids (i.e. > 6 cP). The low viscosity solutions offered less resistance to the integral fountain disintegration process, thereby producing smaller droplets and higher outputs.     

Stability of ricobendazole in aqueous solutions

The chemical stability of ricobendazole (RBZ) was investigated using a stability-indicating high performance liquid chromatographic (HPLC) assay with ultraviolet detection. The degradation kinetics of RBZ in aqueous solution was evaluated as a function of pH, buffer strength and temperature. The oxidation reaction in hydrogen peroxide solution was also studied. Degradation products were analyzed by mass spectroscopy and degradation pathways are proposed. Degradation of RBZ followed pseudo first-order kinetics and Arrhenius behavior over the temperature range 24–55 °C. A V-shaped pH-rate profile over the pH range 2–12 was observed with maximum stability at pH 4.8. The shape of the pH-rate profile was rationalized by catalytic effects of various components in the solution on each RBZ species. At pH 11 the activation energy for hydrolysis was 79.5 kJ/mol, and phosphate catalysis was not observed. Oxidation occurred in hydrogen peroxide solutions and was catalyzed by the presence of copper (Cu²⁺) ions. Ricobendazole amine and albendazole sulfone were identified by MS assay to be the degradation products of hydrolysis and oxidation respectively.     

Photo-isomerization of fluvoxamine in aqueous solutions

The hydrolysis and photolysis of fluvoxamine, a selective serotonin reuptake inhibitor, in aqueous buffer solutions (pH 5, 7, and 9), in synthetic humic water, and lake waters were investigated in the dark and in a growth chamber outfitted with fluorescent lamps simulating the UV output of sunlight at 25 degrees C. No significant hydrolytic degradation/isomerization was observed for 30 days in all aqueous solutions. However, fluvoxamine was moderately isomerized to its (Z)-isomer by simulated sunlight. The photo-isomerization occurred in two stages. The photo-isomerization occurred rapidly within the first 7 days and slowly thereafter with a rate constant of 0.12-0.19 day(-1) for the first stage and 0.04-0.05 day(-1) for the second stage. Photosensitized rate constants in synthetic humic water and in lake waters were approximately 6-7 times faster than that in pH 9 buffer with the rate constants of 1.15-1.34 day(-1) in the first stage. The (Z)-isomer of fluvoxamine was the only product detected in all aqueous solutions and was identified using LC-ESI-MS.     

Ascorbic acid stability in aqueous solutions

Different water purity provokes a great variation of the stability of ascorbic acid and isoascorbic acid solutions. The effect of temperature on ascorbate aerobic oxidation was assessed by means of Arrhenius plots from which thermodynamic parameters were derived. The presence of bovine serum albumin drastically reduces the vitamin oxidation rate regardless of stereoisomerism. On the other hand the interaction with alkaline phosphatase, an enzyme inhibited by preincubation with vitamin C, does not modify significantly the stability in the experimental conditions used.     

Degradation study of thiotepa in aqueous solutions

The degradation of N,N',N"-triethylenethiophosphoramide (thiotepa) in aqueous solutions has been investigated over the pH range 1-14. Samples were analyzed using a high-performance liquid chromatographic system with UV detection. The degradation kinetics were studied as a function of pH, sodium chloride concentration and temperature. The degradation of thiotepa follows pseudo first order kinetics. The pH-log kobs profile shows that thiotepa is most stable in the pH range 7-11. At pH?11 chloride has no influence on the degradation rate. The degradation products were isolated and the structures identified by mass spectrometry. Chloro adducts of thiotepa are generated in the presence of sodium chloride and in acidic medium. In the pH range 7-11 only the mono-chloro adduct of thiotepa could be found. No detectable degradation products were formed at pH?11.     

pKa determination of benzhydrylpiperazine antihistamines in aqueous and aqueous methanol solutions

The pKa1 and pKa2 values of three benzhydrylpiperazine antihistamines, cyclizine (I), chlorcyclizine (II), and hydroxyzine (III), were determined at 24.5 +/- 0.5 degrees by potentiometric titration in aqueous solution to be 2.16 +/- 0.02 and 8.05 +/- 0.03, 2.12 +/- 0.04 and 7.65 +/- 0.04, and 1.96 +/- 0.05 and 7.40 +/- 0.03, respectively. The pKa2 values were also determined by titration in seven aqueous methanol solutions in the range of 11.5-52.9% (w/w) methanol. The apparent dissociation constants of I-III in the aqueous methanol solutions, psKa2, were plotted according to two linear regression equations from which the values in water, p omega Ka2, were extrapolated. The plotted variables were psKa2 versus methanol concentration (%w/w) and psKa2 + log (water concentration, M) versus 1000/epsilon, where epsilon is the dielectric constant of the aqueous methanol solution. The maximum difference between pKa2 and p omega Ka2 was observed in the case of II where p omega Ka2 was 5.23% higher. Statistical analysis of the linear regression data obtained from the plots showed that slightly better accuracy (p less than 0.13) and correlation (p less than 0.16) were obtained, but the precision was essentially equal with both methods. The observed ratio of Ka1/Ka2 in I-III, 2.75 X 10(5)-7.76 X 10(5), was attributed to solven- and space-mediated field effects and electrostatic induction between nitrogen atoms in the piperazine ring.