Eliminating the Interference of Oxygen for Sensing Hydrogen Peroxide with the Polyaniline Modified Electrode

Polyaniline (PANI) has been shown to possess excellent catalytic activity toward oxygen reduction, however, this molecule may interfere with the electrochemical measurement of other targets when using a polyaniline modified platinum (PANI/Pt) electrode. In this study, we have demonstrated the considerable effects of dissolved oxygen on the sensing of hydrogen peroxide with the PANI/Pt electrode. Accordingly, we proposed a strategy to eliminate the influence of dissolved oxygen with oxygen scavengers. Our results indicated that as an oxygen scavenger sodium thiosulfate was very effective in the removal of dissolved oxygen from the sample solution, and had negligible effect on the quantification of hydrogen peroxide when its applied concentration was below 1 mM.

On the other hand, the reduction of oxygen by the conductive PANI film has also been well recognized and investigated intensively [25][26][27][28][29]. Although PANI as a catalyst for oxygen reduction has potential applications in battery and fuel-cell development [27,[30][31][32], oxygen reduction may bring about unexpected interference while electrochemically measuring the above mentioned biomolecules with a PANI modified electrode. For this reason, a routine measurement of H 2 O 2 is normally carried out under an anaerobic environment, and the sample is diluted with a large volume of buffer solution that has been deoxygenated with pure N 2 [33] In order to establish a procedure that is able to eliminate the interference of oxygen on sensing H 2 O 2 , in this study, we investigated the electrochemical response of oxygen with the PANI modified Pt electrode. Our results demonstrated that both air-saturated and oxygen-saturated solution exhibited significant interference on sensing H 2 O 2 , but the dissolved oxygen could be effectively removed by the addition of oxygen scavengers, such as sodium thiosulfate and ascorbic acid. In addition, oxygen scavengers with various concentrations were examined and the recommended concentration was below 1 mM because of the negligible effects on the reduction of H 2 O 2 .

Chemicals
Hydrogen peroxide (35%, v/v) and aniline monomer were obtained from Merck (KGaA Darmstadt, Germany). Sodium thiosulfate and ascorbic acid were purchased from Sigma-Aldrich (Saint Louis, MO, USA). All other reagents used for buffer and standard solution preparation were purchased from various commercial sources and were of analytical grade.

Electrochemical apparatus
A PC-controlled CHI621B electrochemical analyzer (CH Instruments, Austin, USA) was employed to run cyclic voltammetric experiments for electrode preparation and electrochemical measurement. All experiments were proformed in a miniature electrochemical cell using a modified Pt electrode (area: 0.28 cm 2 ) as the working electrode, a platinum wire as the auxiliary electrode, and a Ag/AgCl 3M NaCl electrode as the reference electrode.

Preparation of electrodes
The PANI/Pt electrode was constructed as previously described [22]. On a ceramic plate (area: 2.00 cm 2 ), platinum was sputtered with a shadow mask desired (area: 0.28 cm 2 ) for 10 min on a sputter instrument (JFC-1200, JEOL, Japan). The Pt/ceramic electrode was washed with 3 M NaOH

Electrochemical measurements
The oxidation and reduction of H 2 O 2 on a Pt or PANI/Pt electrode were quantified with cyclic voltammetry in 0.1 M PBS buffer (pH 6.2). The buffer had undergone deoxygenation with highly pure nitrogen for 20 min before a certain amount of H 2 O 2 was added. During the calibration, pure nitrogen gas was gentle purged on the surface of the sample solution to create an anaerobic atmosphere. To investigate the reduction of oxygen on a Pt or PANI/Pt electrodes, the solution was purged with pure oxygen for 30 minutes to reach the saturate concentration. The fresh prepared oxygen scavenger, sodium thiosulfate or ascorbic acid, was added into the solution before the electrochemical measurement.

Oxidation and reduction on a Pt electrode
It is well known that platinum is able to catalyze the electrochemical oxidation of H 2 O 2 [34]. Within our selected potential range between -0.6 to 0.6 V, the addition of H 2 O 2 gave rise to a near 0.6 V anodic response on the bare Pt electrode ( Figure 1A), which resulted from the oxidation of H 2 O 2 vs. Ag/AgCl in 0.1 M phosphate buffer solution (pH 6.2), as shown in reaction (1): Meanwhile, a major cathodic peak between 0.1 ~ 0.2 V was observed, whereas the potential shift slightly towards negative with the increase of H 2 O 2 concentration. Applied the same electrode to an oxygen-saturated solution with the same potential window, the cathodic peak near 0.2 V was also visible with similar peak shift, but not the reduction of H 2 O 2 near 0.6 V ( Figure 1B). Therefore, the cathodic response to H 2 O 2 in Figure 1A was possibly associated with the further reduction of oxygen on the Pt electrode, in which one-step four electron pathway was proposed as reaction (2) [35]: The potential shift in Figure 1A also indicated an increase in the local concentration of oxygen on the Pt electrode. Both anodic and cathodic responses at 0.6 V and between 0.1~0.2 V in Figure 1A were strongly dependent on the H 2 O 2 concentration. The inset of Figure 1A shows the linear correlation between the anodic peak current at 0.6 V and the concentration of H 2 O 2 over the 0~2.5 mM range (sensitivity: 79.11 A·mM -1 ·cm -2 , R 2 = 0.996). However, under our experimental conditions bubbles were observed on the surface of electrode when the H 2 O 2 concentration was higher than 2.5 mM indicating the local concentration of oxygen had exceeded its saturation point.
By narrowing the potential window to -0.6 ~ 0.4 V, the above redox responses to H 2 O 2 near 0.2 and 0.6 V were reduced significantly (Figure 2A). The small cathodic peak near 0.2 V with high concentration of H 2 O 2 was likely associated with the reduction of oxygen that was formed from the partial decomposition of H 2 O 2 on the Pt electrode. Similar cathodic peak was also observed while sensing the oxygen-saturated solution with the potential window of -0.6~0.4 V ( Figure 2B).
Accordingly, the potential window of -0.6~0.4 V was employed for further investigation.

Oxidation and reduction of H 2 O 2 on a PANI/Pt electrode
A PANI film was then electrochemically synthesized on the surface of Pt electrode to form a PANI/Pt electrode. As shown in Figure 3A (red line a), the modified electrode completely suppressed the hydrogen adsorption-desorption redoxs between -0.5 and -0.6 V that was seen in Figures 1 and 2, thereby the PANI film effectively minimizing the background influences from the Pt electrode.
Alternatively, H 2 O 2 was electrochemically reduced with a peak potential centered at -0.32 V (blue line b in Figure 3A) by the following reaction (3): The inset of Figure 3A shows the linear correlation between the cathodic peak current near -0.32 V and the concentration of H 2 O 2 in the range of 0 ~ 2.5 mM (sensitivity: 101.14 A·mM -1 ·cm -2 , R 2 = 0.999). The SEM image indicated the typical sponge-like PANI film ( Figure 3B). Due to the significant surface area provided by the microstructure of PANI film, the major cathodic response was extremely sensitive to the concentration of H 2 O 2 . Therefore, the PANI modified Pt electrode provided better sensitivity and selectivity than a bare Pt electrode, and had potential applications for fabricating H 2 O 2 involved biosensors.

The effect of oxygen on sensing H 2 O 2
Considering the catalytic activity of PANI towards oxygen reduction [29], the electrochemical measurement of H 2 O 2 with PANI/Pt electrode was normally performed under an anaerobic atmosphere with solution that had undergone a deoxygenation procedure. As depicted in Figure 4, a significant cathodic peak at -0.32 V was observed for 2.39 mM H 2 O 2 in 0.1 M PBS solution (pH 6.2) without any deoxygenation procedure (blue curve a in Figure 4). Purging the solution with pure nitrogen gas for ten minutes before the addition of H 2 O 2 decreased the maximum cathodic response by about 1.82-fold (green curve b in Figure 4). In contrast, purging with pure oxygen increased the maximum cathodic response by about 1.80-fold (black curve c in Figure 4). Therefore, the dissolved oxygen may lead to an overestimation of the concentration of H 2 O 2 measured by PANI/Pt electrode. In particular, the observed peak potential for O 2 reduction was almost identical to that of H 2 O 2 reduction under our experimental condition. Accordingly, the reduction of oxygen might go through a two-step four electron pathway on the PANI film, where oxygen was reduced to H 2 O 2 within the applied potential window followed by immediate reduction of H 2 O 2 through reactions (4) and (3), respectively: There was no H 2 O 2 added for (d) (red).

The effect of oxygen scavengers on sensing H 2 O 2
Although purging with nitrogen is a common option to remove dissolved oxygen, it is not always  [36][37][38], both oxygen scavengers with a concentration below 1mM are capable to completely deprive the air-saturated oxygen according to the stoichiometry of the reaction. Meanwhile, sodium thiosulfate will be more efficient because it has less concentration effect on sensing.

Conclusions
Polyaniline, a functional electro-conductive polymer, has been successfully applied in biosensor fabrication to eliminate the background interference, to enhance the electrochemical sensitivity, as well as to immobilize biomolecules and cells [14][15][16][17][18][19][20][21][22][23][24]. However, the interference from dissolved oxygen on sensing of H 2 O 2 with a PANI modified platinum electrode was inevitable, because the electro-state of PANI might be altered by the oxygen reduction on the PANI film [29]. Although purging with nitrogen has been commonly utilized to eliminate the influence of oxygen when sensing H 2 O 2 [33], this is not always feasible in practice. In this study, we further demonstrated that the reduction of oxygen was well catalyzed by PANI film, which might overestimate the electrochemical response to H 2 O 2 . Two oxygen scavengers, such as sodium thiosulfate and ascorbic acid, were examined for their ability to effectively remove the dissolved oxygen. Although H 2 O 2 has been reported to be directly reduced by both oxygen scavengers, our results demonstrated that the reductions were not comparable to the electrochemical reduction catalyzed by PANI film. As a result, both oxygen scavengers had no significant effect on sensing of H 2 O 2 by the PANI/Pt electrode as long as their concentrations were below 1 mM. Moreover, sodium thiosulfate had a less concentration dependent than ascorbic acid while sensing H 2 O 2 , hence, was more suitable for practical applications.