Different analytical methods to quantify residual PEs in the effluent of treated water are reported (Figure 1). For example, size exclusion chromatography (with ultraviolet detection or refractive index) [4], kinetic methods [5], fluorescent probes [6], tannic acid method [7], two-phase titration [8], polarography [9], voltammetry [10], extraction-spectroscopy [11], clay sedimentation rate [12], systems of injection analysis flow [13,14] and colloidal titration [15].

To quantify and identify PEs with high charge density, in solution, Terayama [15] proposed the colloidal titration. This method is based on the stoichiometric reaction between colloidal particles, positively and negatively charged. The end point is frequently detected with colorimetric indicators by visual inspection or spectrophotometric measurements. Several authors have shown that colorimetric detectors are preferred to quantify cationic PEs. For anionic PEs only few indicators work efficiently; however, it is common to use the well known back-titration [16].

The endpoint of cationic or anionic PE titration, without indicator, may be followed by measuring the zeta potential [2,17], conductimetry [18], viscosimetry (minimum viscosity at the point of equivalence) [19], turbidimetry (maximum turbidity at the point of equivalence) [20], fluorimetry [2,22] and potentiometry [16,23].

In this paper the colloidal titration, with spectrophotometry and zeta potential end point detection, was chosen to show that analytical criteria as equilibrium predominance in parallel reactions and interferences of another PE or metallic species could be considered to improve the quantification of CPE even if they seem to be well known methods.

Cationic PEs are usually determined in water by colorimetric methods with OTB (blue/purple). The end point could be detected by visual inspection; however it depends on eye sensibility of each person, thus UV-Vis spectroscopy is a better technique. Kam and Gregory [24] and Zanuttini and Mocchiutti [9] suggested to relate the absorbance diminution at 628 nm as the titrant is added. They propose that the aqueous solution absorbance diminish first by a dilution effect but then the indicator starts to complex with PVSK, appearing a new absorbance point at 509.5 nm (hipsochromic displacement). As established for competing equilibriums the predominance of one reaction over the other one depends on the concentration of each reactive. For a colloidal titration three issues should also be considered, the formation of a solid phase that may interfere the UV-Vis detection, the pH and the ionic strenght. Zanuttini and Mocchiutti [9], for example, avoid the solid phase formation using surfacetants. Kam and Gregory [24] presented studies of pH and ionic strength. However they do not present any informa-

tion concerning the optimization of OTB concentration. In this paper the optimization of OTB dose is one of the main objectives using the Abs at 509 nm vs µmoles OTB. For direct titration of anionic PEs, there are few suitable indicators that show a distinct color change end point; moreover another disadvantage is that the absorption spectrum of the indicator is pH dependent, needing a buffer solutions shortening the pH range for accurate titration. On the other hand, some studies reported indirect colloidal titration for anionic PEs [25]. This method consists of adding a known amount of cationic PE, and back-titrate the excess of cationic PE with the system PVSK/OTB. However, back-titration is tedious and time consuming. The direct titration method with the system PE(+)/PE(−) using the streaming current and streaming potential technique to monitor the progress of the titration continuously is one method currently used to determine anionic PEs [25]. In this paper we use the direct titration method with the system PE(+)/PE(−) without the use of indicators detecting the end point titration by measuring the zeta potential, and even if it requires the understanding of the physicochemical phenomena (Figure 2), it results very simple to quantify the residual PEs in treated water.

Poly(vinyl sulfate) potassium salt (PVSK), molar mass 170 KDa, obtained from Sigma-Aldrich. The negative charges are attributed to sulfate group in each monomer. Polydadmac (OPTIFLOC C-1008) and Flocculant (Trident 27506) that are commercial PEs intended to be quantified in an industrial waste water treatment process. The concentration of polymers, M (mol/L), was expressed on the basis of monomeric unit which indicates the moles of ionic group per liter of the polymer solution.

Toluidine Blue O (OTB) solution was used as an indicator and was obtained from Sigma-Aldrich. Tin solution standard for AAS, 1000 mg/L (FLUKA).

Absorbance of the sample solution was measured at 400 - 800 nm with a spectrophotometer UV-Visible (Cary 100 Conc, Varian) using a 10 mm pathlength cell.

Turbidimetry. The optical density of aqueous polymer solutions was monitored at 500 nm by means of a Uvvisible spectrophotometer (Cary 100Conc, Varian) using a 10 mm pathlength cell.

For zeta potential measurements during titration of cationic PEs the Zetasizer ZS (Malvern) instrument was employed [26].

Synthetic PVSK-OTB solutions were prepared mixing 1.0 mL of a 1.24 mM PVSK and different additions of a 0.27 mM OTB into a 10 mL final volume.

First the optimal OTB/PVSK was studied to detect the equivalence point of the titration according to the following procedure: 1 mL of solution added PVSK 1.24 mM and dilute to 10 mL with deionized water in a volumetric flask 10 mL giving a final concentration of 0.124 mM, this was taken as blank. Other solutions were prepared containing 1 mL of 1.24 mM PVSK and additions

from 0.1 to 1 mL of 0.27 mM OTB and diluted to 10 mL with deionized water. A solution was prepared with 1 mL of 0.27 mM OTB and diluted to 10 mL with deionized water. UV-Vis absorbance of each prepared solutions were done and a graph of absorbance at 509 nm vs µmoles OTB, was constructed.

The determination of cationic PE was made by the colloidal titration method with spectrophotometric detection of the equivalence point using the color change of the OTB (blue to pink) via the PE system/indicator. Indicator solution was prepared by 0.27 mM OTB, a solution of 1.24 mM PVSK which was used as titrant.

A synthetic Polydadmac (10 mg/L) + OTB (13.5 µM) solution was prepared in a volumetric flask of 10 mL. After the UV-Vis absorbance was read, the solution of the cell was returned to a 20 mL vial and added a known amount of titrant PVSK, stirred for 30 s, and returned to the UV-Vis cell for its absorbance measurement again. In the same way, titrant additions continued until the end of titration (the color changed from blue to pink) of the positive PE solution.

According to the colloidal titration method described above, the titration of synthetic Polydadmac solutions in the concentration range of 1 to 100 mg/L using the system PVSK/OTB were carried out until the end point detection.

Polydadmac synthetic solutions were prepared in the concentration range of 1 to 100 ppm, taking a certain amount of a solution 1.28 mM PVSK and diluting to 10 mL with distilled water in a volumetric flask. The prepared solution was poured into a 20 mL vial and the zeta potential was measured initial solution to be titrated. Next was added a certain amount of titrant 1.28 Mm PVSK solution, was stirred for 30 s and measured the pZ of the solution. There have been various additions of titrant until it reached the isoelectric point and the solution turned turbid.

Solutions of 5 ppm Polydadmac in the presence of varying amounts of Sn (1, 2.5, 5 and 10 ppm Sn), then titrated with PVSK detecting the equivalence point by the spectrophotometric method.

Solutions of 5 ppm Polydadmac in the presence of varying amounts of flocculant (1, 2.5 and 5 ppm Flocculant Trident 27506), then titrated with PVSK detecting the equivalence point by the spectrophotometric method.

With a syringe, 1 mL of sample was placed into a port of the zeta potential cell, the sample was injected slowly checking that all air bubbles were removed. Once the sample begins to come out the other port of the cell, the respective plugs were placed. Capillary cell was inserted in the Zetasizer ZS equipment and read the zeta potential value.

Usually the indicator concentration is at least 10^–3 times less concentrated than the titrant but there is no a strict rule to decide it. In this paper we propose a previous optimization of these parameters based on the used method for metal titrations with spectrophotometric detection [27].

As far as me understand there are no published data about the optimum wavelength and PVSK-OTB ratio, previous to perform the titration of cationic PE, the absorbance spectra of a PVSK and PVSK-OTB solutions were obtained. In Figure 4 free OTB has a maximum absorbance at 628 nm and the intensity decreases when OTB interacts with PVSK. The complex formed by the association of these two species presents an absorbance band at 509 nm. In the first solution (0.1 mL OTB solution + 1.0 mL 1.24 mM KPVS solution) the spectrum shows almost exclusively the absorption band corresponding to the OTB-PVSK complex (509 nm). For the rest of solutions the absorption band corresponding to the formed complex increases, but also the absorption band of free OTB in solution indicating that at higher concentration of OTB, PVSK has been saturated.

Figure 4(b) reports the profile of OTB-PVSK complex absorbance, at 509 nm, versus the amounts (µmoles) of OTB added. It is observed a change in slope at 0.5 mL of OTB solution added (13.5 µmoles). This corresponds