The use of ion-selective electrodes in the determination of drug substances

Supramolecular Atropine Potentiometric Sensor

A supramolecular atropine sensor was developed, using cucurbit[6]uril as the recognition element. The solid-contact electrode is based on a polymeric membrane incorporating cucurbit[6]uril (CB[6]) as an ionophore, 2-nitrophenyl octyl ether as a solvent mediator, and potassium tetrakis (4-chlorophenyl) borate as an additive. In a MES-NaOH buffer at pH 6, the performance of the atropine sensor is characterized by a slope of (58.7 ± 0.6) mV/dec with a practical detection limit of (6.30 ± 1.62) × 10⁻⁷ mol/L and a lower limit of the linear range of (1.52 ± 0.64) × 10⁻⁶ mol/L. Selectivity coefficients were determined for different ions and excipients. The obtained results were bolstered by the docking and spectroscopic studies which demonstrated the interaction between atropine and CB[6]. The accuracy of the potentiometric analysis of atropine content in certified reference material was evaluated by the t-Student test. The herein proposed sensor answers the need for reliable methods providing better management of this hospital drug shelf-life while reducing its flush and remediation costs.

Carbon paste electrodes in the new millennium

In this review (with 500 refs), both electrochemistry and electroanalysis with carbon paste-based electrodes, sensors, and detectors are of interest, when attention is focused on the research activities in the years of new millennium. Concerned are all important aspects of the field, from fundamental investigations with carbon paste as the electrode material, via laboratory examination of the first electrode prototypes, basic and advanced studies of various electrode processes and other phenomena, up to practical applications to the determination of inorganic ions, complexes, and molecules. The latter is presented in a series of extensive tables, offering a nearly complete survey of methods published within the period of 2001–2008. Finally, the latest trends and outstanding achievements are also outlined and future prospects given.

Development of membrane selective electrode for determination of the antipsychotic sulpiride in pharmaceuticals and urine

The construction and electrochemical response characteristics of a poly(vinyl chloride) (PVC) membrane selective electrode for the determination of sulpiride (SPD) are described. The sensing membrane comprised an ion-exchanger formed between the protonated drug and tetraphenylborate (TPB(-)) in a plasticized PVC matrix. The influence of membrane composition on the electrode response was studied. The electrode showed a fast, stable and Nernstian response over a sulpiride concentration range (1 × 10(-4) - 1 × 10(-2) M) with a mean slope of 58.4 ± 0.9 mV dec(-1) of concentration, a mean detection limit of 4.2 × 10(-5) ± 1.2 × 10(-5) M, a wide working pH range (2 - 8) and a fast response time (< 15 s). The electrode showed good selectivity towards sulpiride with respect to some inorganic and organic compounds. When the electrode was applied to the determination of sulpiride in pharmaceuticals and human urine, a high percentage of recovery was attained with no need for sample pretreatment procedures because of the lack of interfering matrix effects.

Application of a trazodone-selective electrode to pharmaceutical quality control and urine analyses

A trazodone-selective electrode for application in pharmaceutical quality control and urine analysis was developed. The electrode is based on incorporation of a trazodone-tetraphenylborate ion exchanger in a poly(vinyl chloride) membrane. The electrode showed a fast, stable and Nernstian response over a wide trazodone concentration range (5 x 10(-5)-1 x 10(-2) M) with a mean slope of 59.3 +/- 0.9 mV/dec of concentration, a mean detection limit of 1.8 x 10(-5) +/- 2.2 x 10(-6) M, a wide working pH range (5-7.5) and a fast response time (less than 20 s). The electrode also showed good accuracy, repeatability, reproducibility and selectivity with respect to some inorganic and organic compounds, including the main trazodone metabolite. The electrode provided good analytical results in the determination of trazodone in pharmaceuticals and spiked urine samples; no extraction steps were necessary. Dissolution testing of trazodone tablets, in different conditions of pH and particle size, based on a direct potentiometric determination with the new selective electrode is presented as well.

New ionophores for vitamin B1 and vitamin B6 potentiometric sensors for multivitaminic control

The construction, evaluation and analytical application of potentiometric sensors sensitive to vitamin B1 and vitamin B6 are reported. The solid contact electrodes were produced using beta-cyclodextrins as ionophores in a carboxylated poly(vinyl chloride) support matrix. Near Nernstian slopes (mV/decade) of 51.7+/-0.8, 60.6+/-0.6 and 61.1+/-1.4, within the intervals (M) of 1.0 x 10(-4) to 1.0 x 10(-1), 5.8 x 10(-5) to 1.0 x 10(-1) and 4.3 x 10(-5) to 1.0 x 10(-1) were obtained, for thiamine and pyridoxine I and II prepared membranes, respectively. A pH operational range of 6.5-8.5 for thiamine and 2-4.5 for pyridoxine electrodes was found. Assessment of selectivity coefficients toward a large number of inorganic cations and organic cations usually present in multivitamin formulations revealed good performance. Analysis of vitamins B1 and B6 in complex multivitamin drugs was achieved with recoveries within the intervals of 95.1-99.6% for thiamine and 95.1-102% for pyridoxine. Furthermore, the results enabled by the proposed procedure revealed good agreement with those provided by HPLC.

Capillary electrophoresis with end-column electrochemiluminescence for the analysis of chloroquine phosphate and the study on its interaction with human serum albumin

This paper describes a novel and sensitive method for the estimation of chloroquine phosphate (CQ) using capillary electrophoresis with end-column electrochemiluminescence (ECL) detection. Under the optimized condition, linear calibration curve was obtained for the system over two orders of magnitude with a detection limit of 3x10(-7) M (S/N=3). The relative standard deviations of the ECL intensity and the migration time were 1.4% and 0.05%, respectively (n=6, 5x10(-5) M CQ). Successful separation of chloroquine phosphate, difenidol hydrochloride and clomifene citrate was obtained at pH 7.0. Using the proposed electrochemiluminescence system, a simple method was proposed to study the interaction between chloroquine phosphate and human serum albumin (HSA), and the number of binding sites and binding constant were estimated to be 32.6 and 7.7x10(3) M(-1), respectively.

A novel method for fast determination of Ranitidine in its pharmaceutical formulations by fast continuous cyclic voltammetry

INTRODUCTION

A novel method for the determination of Ranitidine in flow injection systems was developed.

METHODS

Some investigations were also done to find the effects of various parameters on the sensitivity of the method. The conditions producing optimal performance were a pH value of 2, a scan rate value of 100 V/s, accumulation potential of (-100) mV, and accumulation time of 0.4 s. Some of the advantages of the proposed method are as follows: the removal of oxygen from the test solution is not required any more, the detection limit of the method is sub-nanomolar and finally, the method is fast enough for determination of compounds in a wide variety of chromatographic methods. We also introduce a special computer based numerical method, for calculation of the analyte signal and noise reduction. After subtracting the background current from noise, the electrode response was calculated, based on partial and total charge exchanges at the electrode surface. The integration range of currents was set for all the potential scan ranges, including oxidation and reduction of the Au surface electrode, to obtain a sensitive determination. The waveform potential was continuously applied on an Au disk microelectrode (12.5 microm in radius).

RESULTS

The detection limit of the method for Ranitidine was found to be 25 pg/ml. For 8 runs, the relative standard deviation of the method at 1.1 x 10(-8) M was 2.1%.

DISCUSSION

The method was successfully applied for fast determination of Ranitidine in its pharmaceutical formulations. Being very simple, precise, accurate, time saving and economical this method has many advantages compared to all previously reported methods.

Mebendazole Selective Membrane Sensor and Its Application to Pharmaceutical Analysis

A PVC membrane sensor for the selective determination of mebendazole (MBZ) was fabricated. The sensor is based on an ion association of MBZ with silicotungstic acid (STA) as ion pair and bis(2-ethylhexyl)phthalate (BEP) as the plasticizing agent in a PVC matrix. The sensor showed a linear response for MBZ for a concentration range 1.0x10(-6)-5.0x10(-2) M with a Nernstian slope of 55.8 mV/decade (limit of detection 6.3x10(-7) M) in the pH range 4-7. It has a fast response time of <30 s. The sensor showed a very good selectivity for MBZ with respect to a large number of ions. The direct determination of MBZ in pharmaceutical formulations gave results that compare well with the data obtained from the standard method.

Development of in situ ion selective sensors for dissolution

The dissolution of formulations of the drugs dapoxetine, paliperidone, cinnarizine, tetrazepam, mebeverine, loperamide, galantamine and ibuprofen was studied by an in-line potentiometric measurement system. The transpose of a Nikolskii-Eisenman type function performed the conversion of potential to percentage of dissolution. A novel gradient membrane electrode was developed especially for dissolution, varying continuously in composition from an ionically conducting rubber phase to an electronically conducting solid state PVC/graphite composite. The gradient part had a thickness of 200 microm. The electrodes life span exceeded 6 months. An ion exchange procedure was used to prepare them for one specific drug. This enabled us to use one universal electrode built to measure a wide array of drugs. The system parameters such as accuracy, reproducibility and linearity were presented with the data obtained for the drug dapoxetine. In dissolution, accurate measurements were possible from 10(-9) to 10(-3) M concentrations, for high log P drugs. The effect of t(90) response times on the measurement error was estimated. The t(90) response times of the electrodes were concentration dependent, and varied between 50 and 10 s for, respectively, 10(-6) and 10(-3) M concentrations. Potential drift was studied in detail. The measurements performed with these electrodes showed an accuracy of 1%, and inter- and intra electrode variabilities of 0.6 and 1.7%, respectively. The electrodes were successfully applied in colloidal media containing suspended matter, typically formed during dissolution of tablets. The advantages and pitfalls of potentiometry over the presently used techniques for dissolution testing are discussed.

Potentiometric behaviour of ion selective electrodes based on iron porphyrins: the influence of porphyrin substituents on the response properties and analytical determination of diclofenac in pharmaceutical formulations

The potentiometric response characteristics of diclofenac selective electrodes based on Fe(III) tetraphenylporphyrin-chloride (Fe(III)TPP-Cl) and Fe(III) tetrakis(pentafluorophenyl)porphyrin-chloride (Fe(III)TPFPP-Cl) in different mediator solvents and ionic additives are compared. The sensitivity, working range, detection limit, response mechanism, and selectivity of the membrane sensor show a significant dependence on the type of carrier substituent and on the pH value of the sample solution. Studies performed with different amounts of cationic additive (tetra-n-octylammoniumbromide (TOABr)) and anionic additive (sodium tetraphenylborate (NaTPB)) in the membranes allowed the determination of the potentiometric mechanism of action of the used metalloporphyrins. For the analysis of real samples, Fe(III)TPFPP-Cl (type G), prepared in o-NPOE, incorporating 10 mol% of TOABr, was used. This potentiometric unit presented a linear response towards diclofenac concentrations between 10(-5) and 10(-2)mol l(-1) ( I=0.1 mol l(-1)) and slopes of about -59 mV dec(-1), exhibiting a response time of 10s in a buffered solution of ammonia-ammonium sulphate with pH 9.9. The potentiometric analysis of sodium diclofenac in pharmaceutical formulations was carried out by direct potentiometry and the obtained results were compared to those provided by HPLC, presenting relative errors inferior to 1.0%.

Amodiaquine polymeric membrane electrode

The construction and electrochemical response characteristics of two types of poly(vinyl chloride) (PVC) membrane sensors for the determination of amodiaquine hydrochloride (ADQ.2HCl) are described. The sensing membrane comprised an ion-pair formed between the cationic drug and sodium tetraphenyl borate (NaTPB) or potassium tetrakis(4-chlorophenyl) borate (KTCPB) in a plasticized PVC matrix. Eight PVC membrane ion-selective electrodes were fabricated and studied. Several plasticizers were studied namely, dioctyl phthalate (DOP), 2-nitrophenyl octyl ether (NPOE), dioctyl phenylphosphonate (DOPP) and bis(2-ethylhexyl)adipate (EHA). The sensors display a fast, stable and near-Nernstian response over a relative wide ADQ concentration range (3.2 x 10(-6) to 2.0 x 10(-2) M), with slopes comprised between 28.5 and 31.4 mV dec(-1) in a pH range comprised between pH 3.7 and 5.5. The assay of amodiaquine hydrochloride in pharmaceutical dosage forms using one of the proposed sensors gave average recoveries of 104.3 and 99.9 with R.S.D. of 0.3 and 0.6% for tablets (Malaritab) and a reconstituted powder containing ADQ.2HCl, respectively. The sensor was also used for dissolution profile studies of two drug formulations. The sensor proved to have a good selectivity for ADQ.2HCl over some inorganic and organic compounds, however, berberine chloride interfered significantly. The results were validated by comparison with a spectrophotometric assay according to the USP pharmacopoeia.

Construction and evaluation of PVC and sol-gel sensor membranes based on Mn(III)TPP-Cl. Application to valproate determination in pharmaceutical preparations

The construction and general performance of new valproate-selective electrodes based on manganese(III) tetraphenylporphyrin [Mn(III)TPP-Cl], as an ionophore, are presented. The ionophore was incorporated into PVC and ceramic membranes (sol-gel) based on methyltriethoxysilane. The influence of membrane composition and pH and the effect of lipophilic cationic and anionic additives in PVC membranes were investigated concerning their influence on the slope, response time, selectivity and lifetime of the electrodes. The PVC membrane without additive and the sol-gel membrane presented slopes and practical limits of detection of -60.8 mV dec(-1) and 5x10(-6) mol l(-1) and -60.3 mV dec(-1) and 1x10(-4) mol l(-1), respectively. The sol-gel membranes displayed higher selectivity for valproate when compared with PVC membranes. These two types of electrodes were coupled to a sequential-injection analysis (SIA) system for the direct determination of valproate in pharmaceutical formulations. The association of Mn(III)TPP-Cl with the sol-gel support inserted in a SIA system provided potentiometric sensors with an analytical range of 1x10(-3)-5x10(-2) mol l(-1), with a sample rate of 55 samples per hour and a sample and carrier consumption of 140 and 2,500 microl per determination, respectively.

Ion-selective electrodes based on metalloporphyrins for gibberellic acid determination in agricultural products

This work describes the construction, evaluation and analytical application of electrodes selective to the gibberellate anion for the determination of gibberellic acid in agricultural products. Several types of PVC membrane electrodes without internal reference solution were prepared using the manganese(III) complex of meso-tetraphenylporphyrin (TPP) as ionophore and dibutyl phthalate (DBP), as plasticizer. The incorporation of lipophilic chemical species as additives, was also carried out aiming the evaluation of the response characteristics of the electrodes. To accomplish the analysis of commercial agricultural products a selective membrane composed of 28.0% (w/w) of PVC, 66.0% (w/w) of plasticizer and 6% (w/w) of ionophore was used, with no additive. This potentiometric unit presented a linear response between 10(-4) and 10(-1) mol L(-1) in gibberellate, a slope of about -69 mV dec(-1) and a reproducibility of about +/-1 mV day(-1). The potentiometric analysis of gibberellic acid in commercial products was carried out by direct potentiometry and the results obtained were compared with those provided by HPLC.

Polymeric membrane ion-selective electrode for determination of bismuth (III) in pharmaceutical substances

A PVC-based membrane of tridecylmethylammonium chloride reveals a Nernstian potentiometric response (with slope of 55 mV/decade and a correlation coefficient of 0.99) for complex of bismuth (III) with ethylenediaminetetraacetate-anion [Bi(EDTA)](-) over a wide concentration range (1 x 10(-1) to 3 x 10(-5) mol/l). The potential of this electrode is independent of pH in the range of 4.0-11.0. It has a fast response time of about 30 s and was used for a period of 3 months with good reproducibility. The detection limit of this membrane electrode was 1 x 10(-5) M. This sensor exhibits a very good selectivity for [Bi(EDTA)] over a wide variety of complex metal ions with ethylenediaminetetraacetic acid. The proposed electrode has been used in the direct potentiometric monitoring of bismuth (III) in stomach anti-acids.

Flow-injection extraction-spectrophotometric method for the determination of ranitidine in pharmaceutical preparations

The spectrophotometric determination of trace amounts of ranitidine was carried out by liquid-liquid extraction using bromothymol blue with a flow system. The determination of ranitidine in the range of 1 x 10(-5) - 1 x 10(-4) mol l(-1) was possible with a sampling frequency of 40 samples h(-1). The method was satisfactorily applied to the determination of ranitidine in pharmaceutical preparations and the recovery was quantitative and no interferences from excipients were observed.

Multi-task flow system for potentiometric analysis: its application to the determination of vitamin B6 in pharmaceuticals

A flow set-up based on the sequential injection analysis concept was designed, aiming at increased automation and robustness of procedures related with potentiometric detection in pharmaceutical control. In this sense, programmable set-up calibration, ion-selective electrode characterization, standard addition techniques and titration procedures could be carried out without any stock solutions conventional handling or modification on the physical structure of the flow system. Evaluation of a flow-through vitamin B6 selective electrode and its application to routine analysis of pharmaceuticals were selected as models to demonstrate the system potentialities. The system ability to generate in-line calibrating solutions was verified by comparing the results with those obtained with solutions prepared by the manual procedure. The vitamin B6 determination in pharmaceutical products was carried out and in-line performed recoveries gave values within 97.4-103.5%. The system ability to perform titrations was ascertained using the precipitation reaction of vitamin B6 with tetraphenylborate. Other profitable features such as lower reagent consumption with a low effluent generation volume were also achieved.

Flow-injection analysis of dopamine in injections with a periodate-selective electrode

Dopamine determination in pharmaceutical preparations based on its oxidation with periodate (IO(4)(-)) using a new IO(4)(-)-selective electrode under flow conditions is presented. An electrode with a tubular configuration, no internal reference solution, and a PVC (31. 2%) membrane, with metaperiodate bis(triphenylphosphoranylidene)ammonium (1.3%) as ion exchanger and 2-nitrophenyloctylether (67.5%) as mediator solvent, was used. Optimization procedures were directed at potentials versus dopamine readings instead of potential versus the remaining IO(4)(-). This approach was achieved by selecting a 50-cm reactor and an overall flow of 7 mL/min, and injecting 70 microL of dopamine standards in a 3.0 x 10(-4) M IO(4)(-) solution. Under these conditions, a linearity range of 8.0 x 10(-3) to 2.7 x 10(-1) g/L, with a slope of 310.1 +/- 7.4 mV L g(-1) and a reproducibility of +/-0.4 mV, were recorded (n = 8). Interference from common excipients was negligible. Under these conditions, analysis of dopamine injections (n = 12) presenting 200 mg/injection gave average and standard deviation values of 201.0 and 3.3 mg/injection, respectively. A simple and inexpensive flow-injection analysis (FIA) manifold, with a good potentiometric detector, enabled the analysis of 200 samples/h without requiring pretreatment procedures. Comparison with the dopamine injection analysis in the United States Pharmacopoeia monograph showed good accuracy, with a relative deviation of -0.2%.

Mexiletine-sensitive membrane electrode for medical application

Response characteristics of mexiletine-sensitive membrane electrodes based on crown ether and ion-exchanger were examined in a physiological saline in order to find an electrode suitable for determining concentrations of this drug under physiological conditions. Among various crown ethers screened, 4',4"(5")-di-tert-butyldicyclohexano-18-crown6 showed the highest sensitivity to mexiletine in physiological saline containing 0.15 M NaCl and 5 mM 4-(2-hydroxyethyl)-2-piperazineethanesulfonic acid (Hepes) NaOH (pH 7.4). However, the detection limit of 30 microM was 10 times higher than that of 3 microM observed with the electrode based on an ion-exchanger, sodium tetrakis[3,5-bis(2-methoxyhexafluoro-2-propyl)phenyl]borate. Having high selectivity against inorganic cations such as Na+ or K+, the electrode using the ion-exchanger enabled us to determine the level of mexiletine in saliva, the monitoring of which is quite effective for controlling the dose of this drug noninvasively. The mexiletine concentrations determined with the mexiletine electrode compared favourably with those determined by high-performance liquid chromatography which requires an additional procedure to extract mexiletine from saliva.

Tetracycline, oxytetracycline and chlortetracycline determination by flow injection potentiometry

This paper describes tetracycline (TCH), oxytetracycline (OTCH) and chlortetracycline (CTCH) determination by flow injection potentiometry. In the flow system proposed TC samples are inserted in a carrier solution and converged with a Cu(II) solution of known concentration; the Cu(II) decrease due to its complexation with tetracyclines (TC) was monitored. The detector used was a homogeneous crystalline CuS/Ag2S double membrane tubular electrode with increased sensitivity. The present system allows tetracyclines determinations within a 48.1-4.8 x 10(3) ppm for TCH, 49.1-4.9 x 10(3) ppm for OTCH and 51.5-5.1 x 10(3) ppm for CTCH and a precision better than 0.4% for the three TC species. This procedure accomplishes 150-200 samples h(-1) with a Cu(II) consumption of about 13 microg determination(-1).

Cefuroxime selective electrodes for batch and FIA determinations in pharmaceutical preparations

Different cefuroxime selective electrodes, without internal reference solution and comprising PVC membranes, were constructed and evaluated. Membranes were prepared with cefuroxime tetraoctylammonium (A) or cefuroxime bis(triphenylphosphoranylidene)ammonium (B) as ion-exchanger, 2-nitrophenyl octyl ether (X) or bis(2-ethylhexyl)sebacate (Y) as plasticizing mediator solvent and 4-tert-otcylphenol (TOP) as additive. From the comparative evaluation of the described electrodes, membranes comprising 2-nitrophenyl octyl ether, cefuroxime tetraoctylammonium and 4-tert-otcylphenol presented better working characteristics. For these electrodes (type XA-TOP), with a lifetime > 5 months, a lower limit of linear range of 2.8 x 10(-4) M, a practical detection limit of 1.3 x 10(-4) M, a reproducibility of approximately +/-0.6 mV day(-1) and a slope of -50.4 mV decade(-1), under H3PO4/NaH2PO4 solutions (pH 3.5; I =0.1 M), were found. The presence of the additive on the membranes was of crucial importance for the electrodes good characteristics. Interference from sulphate, chloride, nitrate, iodide, cefaclor, cefadroxil, cefazolin and cephradine, on the electrodes behaviour was evaluated. Only a slight interference from nitrate and iodide was recorded, being type XA-TOP electrodes the most selective units. Electrodes with a tubular configuration prepared with type XA-TOP membranes, aiming flow injection analysis, were also constructed. When these tubular potentiometric detectors were evaluated in a double-channel flow injection manifold, with 3.5 pH and 0.1 M ionic strength conditions, significantly better working characteristics than those of the corresponding conventional electrodes, namely higher slopes (-54.6 mV decade(-1)) and better reproducibilities (+/-0.2 mV day(-1)), were found. Both conventional and tubular type XA-TOP electrodes were used for injections analyses by batch and FIA, respectively, presenting low consumption of samples and reagents. Relative error deviations to the reference procedures <3.0% were found.

Ion-selective electrodes for promethazine determinations in pharmaceutical preparations and application to flow injection analysis

The construction and evaluation of PM-selective electrodes with a conventional configuration and without internal reference solution are described. Two different PVC membranes were prepared with PM tetraphenylborate dissolved in 2-nitrophenyl octyl ether (type A) or bis (2-ethylhexyl)sebacate (type B). When the electrodes were evaluated in 0.01 M ionic strength solutions, both presented linear responses from 5 x 10(-5) to 1 x 10(-2) M, slopes close to the theoretical value, practical detection limits of approximately 2 x 10(-5) M, reproducibilities of +/- 0.2 mV/day, and a quick response (< 20 s). Using phosphate buffer solutions (pH 6.0), type B electrodes presented a lower practical detection limit and a better reproducibility than type A. Selectivity against several cations was assessed by the separated solutions method, and the type B electrodes were generally more selective. The type B membrane was also used for the construction of tubular electrodes for the automatic analysis of pharmaceutical products by flow injection analysis (FIA). These tubular detectors, evaluated in a low dispersion FIA system, presented better selectivity and sensitivity than the corresponding conventional ones. Several pharmaceutical preparations were analyzed with both conventional and tubular electrodes. The results obtained by FIA presented relative deviations of < 2% when compared with those obtained from the United States Pharmacopoeia monographs.

Construction and evaluation of PVC conventional and tubular tripelennamine-selective electrodes: their application in pharmaceutical preparations

The construction and evaluation of tripelennamine conventionally-shaped ion-selective electrodes and tubular detectors for the determination of this compound in pharmaceutical formulations are described. Electrodes with conventional configuration have been constructed without an internal reference solution, using several types of immobilized ionic sensors in PVC. The different electrode membranes were prepared by using tripelennamine tetraphenylborate as ionic-exchanger, dissolved in 2-nitrophenyl octyl ether (type A), dibutylphthalate (type B) and bis-(2-ethylhexyl)sebacate (type C) as plasticizer solvents. The general working characteristics of the different types of conventional electrodes were evaluated in tripelennamine solutions, with adjusted ionic strength, showing a linear response in the concentration range of about 4 x 10(-5) - 1 x 10(-1) M and a slope near the theoretical value. The electrodes presented a fast response (> 20 s) and a high reproducibility (> or = 0.2 mV per day). The electrode selectivity in the presence of some interferents, such as sodium, potassium, lithium, ammonium, chlorpheniramine, diphenydramine, promethazine, meclizine and pentazocine, was good, particularly for those whose sensor membrane was prepared with tripelennamine tetraphenylborate dissolved in 2-nitrophenyl octyl ether (type A). Tubular detectors were also prepared using the same sensor membrane and were evaluated in a low-dispersion flow-injection manifold. Under these conditions the detectors presented response characteristics similar to those of the corresponding conventionally-shaped electrodes. The analysis of different pharmaceutical forms (creams, syrups and gels) gave good results with mean recoveries of 99.8-100.6% when the experiments were conducted by direct potentiometry and 99.9-100.4% where the same determinations were conducted by flow-injection analysis with tubular detectors.