Elsevier

Analytica Chimica Acta

Volume 407, Issues 1–2, 29 February 2000, Pages 213-223
Analytica Chimica Acta

An automatic titrator based on a multicommutated unsegmented flow system: Its application to acid–base titrations

https://doi.org/10.1016/S0003-2670(99)00826-0Get rights and content

Abstract

A continuous flow methodology to perform acid–base titrations is described. The titrations are carried out on a multicommutated flow system which simulates batch titration procedures. The titration strategy is based on sequential insertion of increasing titrant and decreasing titrand volumes in a reactor, thus accomplishing complete titration curves. The assessment of the titration end point is similar to that of conventional batch procedures. The theoretical model for the determination of titrand concentration without requiring any calibration process is presented and discussed. The present system was evaluated in vinegar acidity determinations and provided an accuracy better than 3% with a good repeatability (relative standard deviation (RSD) = 2.5%; n = 10) and reproducibility (RSD < 5%). The titration accuracy is time-dependent and has been tested in the 2–10 min range.

Introduction

The selection of either direct or titration analysis for determining chemical species depends upon several factors such as precision, sampling rate and instrumentation available, which are usually more favourable for direct analysis. Nevertheless, the lack of selectivity of detection systems and/or non-existence of standards with the same composition as the samples render titration sometimes the only analytical procedure capable of performing some determinations accurately.

To overcome the difficulties usually associated with manual batch titrations, such as time-consuming procedures, several automatic systems have been proposed. Hence, batch titration automatic systems were implemented [1], [2], in which despite the automation of some titration steps, the work of an operator for washing the titrator vessel before each determination is still required. In other works, the titrations were performed by using a fully automatic apparatus [3].

Aiming at the minimisation of the disadvantages of automatic batch titration, such as complexity and expensive equipment, several authors have proposed continuous flow systems. Titrations hereby performed present a stage at which titrant and sample concentrations are at the stoichiometric ratio of the reaction. This stage is usually found when the flow of one of the solutions (titrand or titrant) is kept stable and the other is varied linearly [4], [5], or by exploitation of the linear concentration gradient of titrant generated by an external gradient chamber [6].

The flow injection systems gave rise to a new titration procedure based on gradient exploitation where the distance (considered as a period of time) between two points of identical amplitude in the same analytical signal was directly proportional to the logarithm of the concentration [7], [8]. Hence, several systems have been proposed and developed, like triangle-programmed flow titrations [9], linear pH-buffering single point titrations [10] and coulometric flow injection analysis [11]. The application of automatic burettes for the insertion of different volumes of titrant in a continuous sample flow enabled to attain analytical signals that match the profile of the titration curve [12].

One particular feature of the existing flow titration systems is that a calibration step is required. However, the lack of adequated standard samples for chemical determinations may be a serious limitation.

Strategies based on concentration gradients with a single standard calibration have also been presented [13]. The concentration of one of the solutions (titrand or titrant) is kept at its steady state throughout the whole process by pumping it continuously. The other is injected and its concentration is determined by the gradient calibration technique.

Other works refer to flow titration based on unsegmented [14] or segmented [15] multicommutated flow systems using a binary search strategy for the end-point determination. Although these titrations do not require any calibration, they can only be applied to systems in which an indicator (or other reference solution) can be used by the binary searching algorithm to decide the next step in the searching process.

In this work, a new titration strategy based on the sequential insertion in a reactor (mixing chamber) of increasing volumes of titrant and decreasing volumes of titrand is presented, the inserted volumes being determined by the valve commutation times. This system enables to attain complete titration curves similar to those of batch titration systems. The software developed is able to control every step of the titration procedure, perform data acquisition and processing. The titration end-point was determined by the second derivative method as in batch titration systems. The theoretical model used was tried out and proved to be suitable for the description of the analytical process. The flow system proposed allows to simulate batch titration procedures without requiring a calibration step. The automatic flow titrator developed was tested for acidity determination in vinegar samples.

Section snippets

Titration strategy

The strategy used is based on the sequential insertion in a reactor (a mixing chamber) of increasing volumes of titrant and decreasing ones of titrand, the inserted volumes being determined by the valve commutation times, to attain complete titration curves. Using this strategy, and considering that the flow-rate is constant, the variable volume can be obtained from the time values.

The titrant valve (V) presents two inlets, one (a) for the titrand (S) and another (b) for the titrant (T) and one

Model used

As already referred to, it was assumed that the reactor (mixing chamber) is initially full of titrand. The variation (reduction) of its concentration during the insertion of titrant (Fig. 1B, Step 2) can be expressed by [18]Ca(tx)=C0tn+C0tn+Ca(ty1)expFV(t1−t0)where Ca(tx) is the variation of the titrand concentration (mol/l) in the mixing chamber while the titrant is being inserted; C0t, the initial concentration (mol/l) of titrant; Ca(ty1), the titrand concentration (mol/l) in the mixing

Reagents and solutions

Analytical-reagent grade chemicals and deionized water with a specific conductivity of less than 0.1 μS/cm were used throughout.

Acetic and hydrochloric acid solutions used as standards were prepared from the corresponding standards by adequate dilution. The concentration of the solutions hereby obtained was determined by potentiometric titration with sodium hydroxide solution using a conventional glass electrode. This determination was performed with an automatic batch titrator.

When required,

Flow system manifold

A schematic representation of the flow system developed is shown in Fig. 3. An automatic microburette controlled by a microcomputer and placed at the end line of the system was used to keep a constant flow. A pneumatically actuated valve, whose commutation times were controlled by a microcomputer, was used for the insertion of different titrand and titrant volumes in the system, thus establishing a direct relationship between commutation times and volumes. The different titrand and titrant

Conclusions

The titration system developed allowed titrations similar to those obtained with batch titration systems. This system enabled to obtain complete titration curves, and a determination of the end point by the second derivative method like in batch titration systems. The evaluated theoretical model was appropriate for describing the analytical process and was used for the determination of the titrand concentration determination without requiring any previous calibration process.

As this system does

Acknowledgements

The authors are grateful to the projects AMOCO (ERB-FAIR-CT96-1198) and CNPq/JNICT (910155/96-8). One of us (C.M.N.V.A.) is grateful to JNICT for the Ph.D. grant. We also thank Antonio Conceição (IST) for the discussion about dynamic models.

References (18)

  • L. Pehrsson et al.

    Talanta

    (1977)
  • S.M. Abicht

    Anal. Chim. Acta

    (1980)
  • J. Ruzicka et al.

    Anal. Chim. Acta

    (1977)
  • A.U. Ramsing et al.

    Anal. Chim. Acta

    (1981)
  • B. Fuhrmann et al.

    Anal. Chim. Acta

    (1993)
  • O. Aström

    Anal. Chim. Acta

    (1979)
  • R.H. Taylor et al.

    Talanta

    (1992)
  • M. Korn et al.

    Anal. Chim. Acta

    (1995)
  • P.B. Martelli et al.

    Anal. Chim. Acta

    (1999)
There are more references available in the full text version of this article.

Cited by (22)

  • Novel approaches to analysis by flow injection gradient titration

    2007, Analytica Chimica Acta
    Citation Excerpt :

    Most of the above methods exploit titration curves with a feature indicating the end-point of titration. Among them traditional titration curves showing signal-volume of titrant added dependences [4–6], signal-time [1,2,7–10] or signal-titrant flow rate relationships [3] can be distinguished. Ingenious concepts present propositions applying the binary search approach [11,12] and a one-dimensional optimisation algorithm [13].

  • Multicommutated generation of concentration gradients in a flow-batch system for metronidazole spectrophotometric determination in drugs

    2004, Analytica Chimica Acta
    Citation Excerpt :

    To implement automatic titration in flow, several strategies were described [3–24] and those requiring calibration based on the titration of synthetic samples [3–13] present the following drawbacks: (a) difficulty in preparing the synthetic samples, e.g., in the determination of hardness, oxidability or acidity/alkalinity in natural waters, might limit the applicability of these systems; (b) results may be dependent of more than one analyte concentration, and different analytes may present different reaction stoichiometries towards the titrant, so that preparation of the standard solutions may become cumbersome. Among the proposed strategies that do not request calibration based on the titration of synthetic samples [14–24], there are two where the automatic titration was implemented of a fast way when concentration gradients generated in flow injection systems were exploited [17,20]. However, these systems present a considerable limitation to analyze samples with high variability in the analyte concentration due to necessity of modification in the flow manifold to adjust the adequate relation of instantaneous concentration between sample and titrant at the detection point.

View all citing articles on Scopus
View full text