Ferrocene-enhanced polyvinyl chloride-coated ...

Journal of Electroanalytical Chemistry 760 (2016) 158–164

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jelechem

Ferrocene-enhanced polyvinyl chloride-coated electrode for the potentiometric detection of total residual chlorine in simulated ballast water Xiao-Hui Dai a, Juan Zhang a, Xin-Jing Pang a, Jun-Ping Zhou a, Guang-Zhou Liu b,⁎, Shu-Yong Zhang a,⁎ a b

School of Chemistry and Chemical Engineering, Shandong University, Jinan 250199, P. R. China Sunrui Marine Environment Engineering, Co. Ltd, Qingdao 266101, P. R. China

a r t i c l e

i n f o

Article history: Received 18 August 2015 Received in revised form 15 November 2015 Accepted 24 November 2015 Available online 2 December 2015 Keywords: Ballast water Electrolytic treatment Total residual chlorine Ion-selective electrode Ferrocene

a b s t r a c t In-situ electrolytic generation of chlorine is commercially adopted as an economic and effective method for treating ballast water (BW). An in-situ and rapid method for detecting total residual chlorine (TRC) is necessary for automatic control on the electrolysis. A polymer-coated electrode fabricated by coating a glassy carbon (GC) electrode with a polyvinyl chloride (PVC) coating that contains zephiran chloride (ZephCl) was prepared in this study. This all-solid ion-selective sensor was used to detect TRC in the simulated electrolytically treated BW. Results showed that adding ferrocene (Fc) in the PVC coating could significantly improve the response speed, reproducibility and stability of the electrode. The potential of the Fc–PVC–ZephCl–GC electrode was linearly proportional to the logarithm of TRC within the range of 1 mg/L to 20 mg/L. The interference of the dissolved oxygen and the possible co-existing ions in seawater was also evaluated without evident interference found. The response mechanism of the electrode, the enhancement of Fc and the failure mechanism of the electrode were discussed. The Fc–PVC–ZephCl–GC electrode could be used as a suitable sensor for the in-situ monitoring of TRC in the electrolytic treatment of BW. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Considerable attention has been focused on ballast water (BW) treatment to reduce the ecological risks possibly initiated by discharging BW that contains invasive marine species into new environments [1–4]. Among the developed methods, electrolytic treatment by in-situ generation of chlorine [4–7] is adopted as one of the most economic and effective technologies. During electrolysis, chlorine formed on anode reacts with base generated on cathode: Cl2 þ 2NaOH ⇌ NaCl þ NaClO þ H2 O

ð1Þ

Part of NaClO undergoes hydrolysis: NaClO þ H2 O ⇌ HClO þ NaOH

ð2Þ

Therefore, Cl2, NaClO, and HClO are all effective for sterilization in the electrolytically treated BW. The total concentration of these effective species is summed up as the total residual chlorine (TRC). Based on the equilibria shown in Eqs. (1) and (2), once the concentration of one species, such as ClO−, is determined, the concentration of other species can also be found. This condition indicates that identifying the relationship between the concentration of ClO− and TRC is possible. Results show that an optimal concentration of TRC exists ⁎ Corresponding authors at: School of Chemistry and Chemical Engineering, Shandong University, Jinan 250199, P. R. China. E-mail address: [email protected] (S.-Y. Zhang).

http://dx.doi.org/10.1016/j.jelechem.2015.11.036 1572-6657/© 2015 Elsevier B.V. All rights reserved.

in the electrolytically treated BW for sufficient sterilization [5]. When TRC is less than 5 mg/L, the efficiency of sterilization is insufficient [8]. When TRC is higher than 15 mg/L, the strong oxidation of TRC may harm the marine species that live in the discharging area, damage the ship painting, and cause serious corrosion of the ship steel beneath the painting [9–11]. Therefore, TRC should be maintained within the optimal range of 8 mg/L to 12 mg/L. A timely feedback based on an in-situ sensing of TRC is required to maintain this range. That is, once TRC in BW becomes less than 8 mg/L, the sensor can provide an electric signal to a control unit to trigger the electrolysis equipment to produce extra chlorine. When TRC in BW becomes higher than 12 mg/L, the sensor can automatically send an electrical message to the control unit to turn off the electrolysis current. To date, many methods have been developed to detect TRC, including colorimetric methods based on N, N-diethyl-p-phenylenediamine [5,12] or o-toluidine [13], iodometric titration [14], optical methods [15], and electrochemical methods based on amperometric measurement [16–21]. Although these methods are applied in analyzing tap water, swimming pool water and wastewater, they are all ex-situ methods with their measuring procedures usually being rather complicated and time-consuming. However, these methods cannot directly provide electric responses and fulfill the demand of automatic feedback that triggers the electrolysis equipment as expected. Given these reasons, the potentiometric method based on ion-selective electrodes (ISEs) has attracted the interest of many researchers. This method is a

X.-H. Dai et al. / Journal of Electroanalytical Chemistry 760 (2016) 158–164

more suitable method to monitor TRC in the electrolytic treatment of BW for its easy preparation and operation, low-cost measurement, and short response time. Numerous ISEs such as acetylcholine sensitive enzyme [22], Pb[II]ISE [23], Fe(III)/Fe(II) redox buffer [24], and potassium iodide redox electrodes [25,26], have been fabricated to detect TRC. However, these electrodes can only provide indirect potential response to TRC after several complicated sample preparation procedures. Adding a special redox buffer and adjusting the pH of the solution make these methods unsuitable for detecting the TRC in BW of massive volume. Motomizu et al. [27] first reported the use of coated wire electrode (CWE) in analyzing free chlorine. But the stability and reversibility of the electrode rapidly deteriorate due to the blocked effect of the PVC coating layer and the formation of water layer between the electronic and ionic conductors after a long immersion. Thereafter, some ion-to-electron transducer intermediate layer is introduced in the polymer coating of the CWE electrode to enhance performance. Conducting polymer [28–31], carbonaceous materials [32–34], and redox-active self-assembled monolayer [35] have been tested to improve both the stability and reversibility of the electrode. The results indicate that these all-solid state ISEs are more suitable for in-situ sensing. Polyvinyl chloride (PVC)-coated glassy carbon (GC) electrode with zephiran chloride (ZephCl, also named benzyldimethyltetradecylammonium chloride) dissolved in the PVC matrix was tested in this study for in-situ detection of TRC. This electrode failed to provide satisfactory results. Ferrocene (Fc) was then introduced as the phase transfer catalyst and ion-to-electron transducer into the PVC coating. Fc was considered because of its high redox activity, good electrochemical reversibility and stable redox potential [36,37]. This study showed that with Fc serving as the phase transfer catalyst and ion-to-electron transducer the performance of the PVC-coated GC electrode significantly improved. The response of the Fc-enhanced electrode to the change in TRC was rapid, and the TRC of the simulated BW could be monitored for nearly one week using one electrode. 2. Experimental 2.1. Reagents and materials Based on Eqs. (1) and (2), the effective species in electrolytically treated BW are formed by the reaction of Cl2 with NaOH. To our knowledge, the commercial sodium hypochlorite is prepared by purging Cl2 into the NaOH solution, following the same procedure. The effective components of the electrolytically treated BW are similar to those of the sodium hypochlorite solution with differences in NaCl and TRC concentrations. Hence, the electrolytically treated BW can be simulated by diluting the sodium hypochlorite solution with higher TRC with 3.5 wt.% NaCl solution. A series of solutions containing different TRCs was prepared by dissolving the sodium hypochlorite (Sinopharm chemical Co. Ltd., China) solution in 3.5 wt.% aqueous NaCl solution. Owing to the inaccurate concentration of the commercial sodium hypochlorite, iodometric titration was used to calibrate the concentration of hypochlorite in these solutions. ZephCl was purchased from Aladdin Industrial Co. Ltd. (Shanghai, China). PVC powder, Fc, tetrahydrofuran (THF), Na2SO4, NaF, MgCl2, Na3PO4, Na2CO3, NaHCO3, NaNO2, NaBr, KI, KCl, CaCl2, and NH4Cl are all analytical reagents. All solutions were prepared with ultrapure water (Millipore, 18.25 MΩ∙cm). 2.2. Electrode preparation The solution for coating casting of the PVC–ZephCl–GC electrode was prepared by dissolving 71 wt.% PVC and 29 wt.% ZephCl in 10 mL of THF with an ultrasonic cleaner for 15 min. The solution for the

159

Fc–PVC–ZephCl–GC electrode was prepared by dissolving 52 wt.% PVC, 32 wt.% ZephCl, and 16 wt.% Fc in 10 mL of THF. Prior to casting, a GC electrode (4 mm in diameter) was carefully polished and rinsed with ethanol and ultrapure water using an ultrasonic cleaner for 1 min and then dried using a N2 flux. The PVC coating was casted using a dipping method. After drying in air overnight, the PVC–ZephCl–GC and Fc–PVC–ZephCl–GC electrodes were obtained. Before each measurement, the electrode was activated by immersing in 3.5 wt.% NaCl solution for 30 min. The composition of 3.5 wt.% NaCl is close to that of the simulated electrochemically treated BW. After conditioning in this solution, the electrode will not be sensitive to Na+ and Cl−, which are abundant in seawater. This condition can stabilize electrode potential and shorten response time. After each experiment, the electrode was rinsed with deionized water and stored in 3.5 wt.% NaCl solution. 2.3. Potentiometric measurements The electrochemical performance of the PVC–ZephCl–GC electrode with or without Fc addition was characterized using a two-electrode cell with the PVC–ZephCl–GC electrode or Fc–PVC–ZephCl–GC electrode serving as the working electrode and a saturated Ag/AgCl electrode (REX 217-01) with the KCl inner solution and a 3.5 wt.% NaCl solution salt bridge serving as the reference electrode. For electrochemical impedance spectroscopy measurement, a three-electrode cell with a Pt foil of 1 cm2 serving as the counter electrode was used. All the potentials reported in this work were referred to this Ag/AgCl reference electrode. The electrolyte solutions are 3.5 wt.% NaCl solution containing different TRCs. The concerned species were dissolved in the 3.5 wt.% NaCl solution with their highest concentration set close to that of seawater to investigate the interference of the dissolved oxygen and the possible co-existing ions in seawater. The potential response of the electrode to TRC was recorded by measuring the open circuit potential (OCP) using the two-electrode cell on a CHI 604C electrochemical workstation (CH Instrument, Shanghai, China). All measurements were conducted at room temperature (25 ± 0.2 °C). Electrochemical impedance spectroscopy of the electrode was recorded using the three-electrode cell at OCP. The measuring frequency range is 0.01 Hz to 105 Hz with the excitation signal of 50 mV. The impedance was analyzed by fitting the spectrum using the ZsimpWin impedance analysis software. 3. Results and discussion 3.1. Response behavior The potential responses of the PVC–ZephCl–GC electrode and Fc–PVC–ZephCl–GC electrode to the TRC change are shown in Fig. 1. The electrode potential for the PVC–ZephCl–GC electrode could not attain a stable value, and no reliable result can be obtained. By contrast, the potential of the Fc–PVC–ZephCl–GC electrode could attain a stable value in less than 50 s. This result indicates that the adding Fc could significantly improve the response behavior of the electrode. Fig. 1 shows that the electrode potential of the Fc–PVC–ZephCl–GC electrode is linearly proportional to the logarithm of TRC (logc), obeying Nernst equation. However, the slope of the linear part is 67.6 mV/decade, which deviates from the typical Nernstian slope of 59.16 mV/decade. The stability of the Fc–PVC–ZephCl–GC electrode was evaluated by performing the measurement using the same Fc– PVC–ZephCl–GC electrode in four days. After each measurement, the electrode was rinsed with deionized water and stored in 3.5 wt.% NaCl solution. The deviation of the electrode potential during four days was not evident, thereby demonstrating good electrode stability. The small potential deviation can be ascribed to the relatively low sodium hypochlorite concentration and the small memory effect of the electrode (vide infra).

160


Fig. 1. Potential-dependence of Fc–PVC–ZephCl–GC electrode on TRC in 3.5 wt.% NaCl solution. The inset shows the variation of electrode potential during each measurement.

3.2. Response speed and potential recovery Response time is defined as the interval required for the variation in electrode potential becoming less than 1 mV/min. Based on the inset of Fig. 2, the OCP of the Fc–PVC–ZephCl–GC electrode became stable in less than 50 s. The respond speed of the electrode to the change in TRC is sufficiently rapid for performing practical measurement. The response reversibility of the electrode was also evaluated by transferring the Fc–PVC–ZephCl–GC electrode from the 3.5 wt.% NaCl solution containing 5 mg/L TRC and that containing 15 mg/L TRC. The electrode potential rapidly recovered the initial value after the transfer (Fig. 2), showing good reversibility and potential recovery.

3.3. Electrochemical impedance spectroscopy The impedance spectra of the PVC–ZephCl–GC and Fc–PVC– ZephCl–GC electrodes in the 3.5 wt.% NaCl solution containing different TRCs are shown in Fig. 3.

Fig. 2. The potential recovery of the Fc–PVC–ZephCl–GC electrode after transfer between 3.5 wt.% NaCl containing 5 mg/L and 15 mg/L TRC. The inset shows the respond time.

Fig. 3. Nyquist plots of the PVC–ZephCl–GC electrode (a) and Fc–PVC–ZephCl–GC electrode (b) in 3.5 wt.% NaCl containing different TRCs.

Fig. 3(a) shows that for the PVC–ZephCl–GC electrode, the impedance spectra recorded in 3.5 wt.% NaCl solutions containing different TRCs coincide with each other, demonstrating the poor sensitivity of the PVC–ZephCl–GC electrode to the change in TRC. Hence, PVC– ZephCl–GC electrode is not suitable for TRC detection. Fig. 3(b) shows that for the Fc–PVC–ZephCl–GC electrode, the impedance of the electrode significantly changes with TRC, therefore suggesting the possibility for monitoring TRC change using this electrode. The Nyquist plots of both PVC–ZephCl–GC and Fc–PVC–ZephCl–GC electrodes are composed of two semicircles in higher frequency and a straight line in lower frequency. The insets of Fig. 3 are the higher magnification spectra of high frequency range. In the high frequency range of internal illustrations, the intersection of the impedance spectrum with the Z′ axis is related to the solution resistance (Rs), and the first semicircle with small diameter is related to the charge-transfer resistance (Rct) of the GC/PVC interface. Based on the comparison of the two insets in Fig. 3(a) and (b), adding Fc can reduce Rct, thereby suggesting an increase in exchange current density (iex). The lowered charge-transfer resistance and increased exchange current density can shorten the response time and improve the potential stability of the electrode. The second semicircle in Fig. 3 corresponds to the resistance of the PVC coating (Rc). The diameter of this semicircle also decreases with increasing the TRC. This result suggests that the transport of ClO− anions into the PVC


membrane (Γ) from the aqueous solution becomes easier at higher TRC because of the higher concentration gradient. This can be learned from Fick's first law for diffusion: Γ¼DA

dc dx

ð3Þ

The equilibrium constant of this equilibrium can be expressed as follows: h

i h i − − − − ClOcoating Clsln ClOcoating Clsln h i h i K¼ ¼ − − − − ClOsln Clcoating ClOsln a− ClOcoating

ð6Þ

T

The higher mobility of ClO− in PVC coating also facilitates the transport. With the increase of TRC in the film, the exchange current density (iex) at the GC/PVC interface is also increased based on the following equation [38]: iex ¼ nFK c;0 cox ð0; t Þ

ð4Þ

The increased i ex can shorten the time for the electrode to attain electrochemical equilibrium and stabilize the electrode potential. The impedance was also analyzed by fitting the spectra based on the equivalent circuit shown in the inset of Fig. 3(b). In the equivalent circuit, Qc and Qdl are the constant phase elements relevant to the PVC coating and GC/PVC coating interface, respectively, and Zw is the Warburg impedance relating to mass transport from the bulk solution to the electrode surface. Table 1 shows that with increasing the TRC, Rct and Rc of the electrode both decrease. The decrease in Rct manifests that the redox reaction ClO− + H2O + 2e− ⇌ Cl− + 2OH− accounting for the potential establishment at the GC/PVC interface is accelerated, whereas the decrease in Rc reflects the accelerated transport of ions through the PVC coating. 3.4. Theoretical discussion of the response mechanism As experimentally demonstrated in the above section, the addition of Fc in the PVC coating can significantly enhance the response of the PVC–ZephCl–GC electrode to the TRC change. The presence of ZephCl is crucial for the response of the PVC-coated electrode to respond to TRC [27]. The effect of ZephCl can be explained by considering the interaction between Zeph+ and ClO−. The interaction between Zeph+ and ClO− is considerably larger than that between Zeph+ and Cl− [39,40] because of the larger radius of ClO−. Theoretically, for the same cation (Zeph+) with the same electric field strength, the induction effect of the cation exerting on anion depends on the radius of the anion. Larger anion size results in stronger induction effect. The interionic attraction between Zeph+ and ClO− is stronger being as ClO− is larger. The stronger interaction between Zeph+ and ClO− makes the entrance of ClO− to the PVC coating easier, resulting in an enrichment of ClO− in the PVC coating. At stable conditions, equilibrium is established between ClO− in the PVC coating and in the NaCl solution: −

161

−

−

−

ClOsln þ Clcoating ⇌ ClOcoating þ Clsln

ð5Þ

Table 1 Fitting results of the Fc–PVC–ZephCl–GC electrode in 3.5 wt.% NaCl with different TRCs. TRC/mg L−1

0

1

5

10

15

20

Rs/Ω 104 × Rc/Ω 105 × Qc/S·sn n Rct/Ω 106 × Qdl/S·sn n 105 × Zw/S·s5

26.6 27.84 1.401 0.8532 2089 3.85 0.9165 0.3333

26.88 26.02 1.371 0.8451 2000 4.245 0.9143 0.335

27.1 14.01 1.133 0.8429 359.2 3.662 0.9309 2.389

24.27 8.396 1.516 0.8124 174.8 3.584 0.9317 5.001

27.05 5.848 2.012 0.7882 62.92 1.773 0.9946 10.18

25.73 4.496 2.471 0.7722 60.65 1.603 1.000 14.88

Where a represents the total concentration of ion-exchange sites of Zeph+ in the coating membrane, and (a − [ClO− coating]) is in fact the ionexchange site in PVC film occupied by Cl−. According to the Nernst equation, the electrode potential of the PVC-coated electrode containing oxidized Fc+ in the PVC–ZephCl coating can be written as follows:

φcoating

h i Fcþ coating RT i ln h ¼ φ⊖ þ F Fc0coating

ð7Þ

The reaction of ClO− with Fc0 is −

−

ClO þ 2Fc0 þ H2 O ⇌ Cl þ 2Fcþ þ 2OH−

ð8Þ

If the reaction completes, then h

i h i − Fcþ coating ¼ 2 ClOcoating

ð9Þ

Substituting Eqs. (6) and (9) into Eq. (7) gives: φcoating ¼ φ⊖ þ

− 2aK ClOsln RT ln − − F Clsln þ K ClOsln

ð10Þ

− Because in solution, [Cl− sln] is much higher than [ClOsln], so that this equation can be reduced as

− 2aK ClOsln RT − ln F Clsln RT RT − RT − ⊖ ln 2aK− ln Clsln þ ln ClOsln ¼φ þ F F F

φcoating ≈ φ⊖ þ

ð11Þ

Owing to the high concentration of Cl− in 3.5 wt.% NaCl solution, [Cl− sln] remains nearly unchanged after the exchange reaction (5). Therefore, the increased ClO− sln in the solution can finally result in a corresponding increase in electrode potential. The second term in Eq. (11) also results in an increase in electrode potential. Although this term cannot change the slope of the Nernst equation, it can increase the electrode potential and reduce the relative error of the potential measurement. The time for the electrode to establish a stable potential response to TRC can also be considered as the time required for the exchange reaction shown in Eq. (5) to attain its equilibrium. Our discussion is different from that proposed by Fu et al. [41]. Given that the electrode potential requires considerable time to attain real electrochemical equilibrium, the authors considered the response potential of the polymer-coated electrode a quasi-steady state response. They ascribed the potential response to the quasi-steady flux of polyion to the electrode surface and into the coating matrix. The above mentioned experimental results demonstrate the significant improvement in the response rate of the PVC–ZephCl–GC electrode by adding Fc in the PVC–ZephCl coating. This effect can be attributed to the formation of Fc+ caused by the prior oxidation of Fc during storage, as has been revealed by the voltammetric measurement and the oxidation of Fc caused by ClO− in the solution as predicted by some other authors [42]. Theoretically, to maintain charge neutralization, the formation of Fc+ in the PVC − ZephCl coating requires the diffusion of extra anions from the solution into the PVC coating. The solution contains two types of anions, namely, Cl− and ClO−. Owing to the stronger interaction between Zeph+ and ClO−, the formation of Fc+ would spontaneously facilitate the diffusion of ClO− into the PVC coating. Therefore,

162


the formation of Fc+ can accelerate the diffusion of ClO− and shorten the time for the exchange reaction shown in Eq. (5) to attain equilibrium. The rapid approach to equilibrium may further shorten the response time. Fig. 4 depicts the potential response of the Fc–PVC– ZephCl–GC electrode to TRC. The interface between the aqueous phase and the coating film is the interface where the electrolytic extraction of ClO− occurs. When ClO− is extracted in coating membrane, the redox reaction of the Fc on the GC electrode occurs with the simultaneous doping-dedoping of Cl−. 3.5. Possible interference 2+ , and BW contains numerous cations, such as K +, NH+ 4 , Mg 2− 2− − − Ca2 +, as well as anions, such as Br−, I−, NO− , SO , CO , HCO 2 4 3 3 , F , and PO3− 4 . Whether or not these ions interfere with the measurement of TRC should be evaluated. Owing to the redox properties of Fc, the dissolved oxygen in BW may also interfere with the TRC measurement. The most common approach to eliminate the interference of dissolved oxygen is purging the solution with inert gas such as Ar and N2. But gas purging is not suitable for the simulated electrolytically treated BW because free chlorine can also be removed from the solution with the gas flux. Meanwhile, purging the entire BW with pure Ar or N2 is impossible for the practical application of the TRC detection in BW. Therefore, the effect of dissolved oxygen on the TRC detection must be evaluated and managed to pre-exclude its effect. To evaluate the interference of the dissolved oxygen, blank tests were performed by replacing the sodium hypochlorite solution with distilled water that is purged with Ar to remove the dissolved oxygen prior to using of the same volume in preparing the 3.5 wt.% NaCl solution with different TRCs. The interference of the possible co-existing ions was evaluated by adding these ions into the simulated BW. The effect of dissolved oxygen on the electrode potential is quite small compared with the significant potential change with the increasing TRC, which is shown in Fig. 5. This small interference can be caused by the low concentration and less oxidation of the dissolved oxygen with regard to sodium hypochlorite. Therefore, neglecting the interference of the dissolved oxygen is reasonable. A method was used to evaluate the interference of the possible co-existing ions. The possible interference was studied in 3.5 wt.% NaCl solution that contains only one possible co-existing ion. The highest concentration of the possible interfering ions was adjusted to the value close to that in BW. Fig. 5 illustrates that although the effect of these possible ions on the TRC measurement is different, their

Fig. 4. Schematic of the potential response of Fc–PVC–ZephCl–GC electrode to TRC.

Fig. 5. Potential-dependence of the Fc–PVC–ZephCl–GC electrode on TRC in 3.5 wt.% NaCl solution and the interference of dissolved oxygen and co-existing ions.

interference is less dependent on their concentration. Except for 3.5 wt.% NaCl solution that contains 1 mg/L TRC, the interference of these ions is relatively small. Given that the optimal concentration of TRC for sterilization is between 8 and 12 mg/L, the interference of the possible co-existing ions can be neglected. 3.6. Stability of the electrode The stability of the Fc–PVC–ZephCl–GC electrode was evaluated by measuring the electrode potential of the same electrode in 3.5 wt.% NaCl that contains 10 mg/L TRC after storage in 3.5 wt.% NaCl for different days. Fig. 6 indicates that the electrode potential nearly remains unchanged in the previous six days. From the seventh day onward, potential deviation became larger with prolonging the storage. After two weeks of storage, the electrode potential failed to attain a stable value and therefore failed in the TRC response. The failure of the electrode was attributed to three different causes: the dissolution of Zeph+ and Fc+ in the simulated BW, the memory effect of the electrode, and the diffusion of water to the GC/PVC interface forming a water layer.

Fig. 6. Potential variation of the same Fc–PVC–ZephCl–GC electrode in 3.5 wt.% NaCl containing 10 mg/L TRC measured in different days.


Zeph+ and Fc+ play important roles in the potential response of the Fc–PVC–ZephCl–GC electrode. During immersion in the NaCl solution, the gradual and avoidless dissolution of Zeph+ and Fc+ deteriorates the performance of the electrode [43–45]. After each measurement, some ClO− anions may be trapped in the PVC coating, resulting in a gradual increase in electrode potential. This phenomenon is known as the memory effect of the electrode, which can also worsen the potential recovery of the electrode. Due to the polar nature of PVC and the hydrophilic nature of ZephCl, water can diffuse into the PVC coating and reach the GC/PVC interface, thereby forming a water layer. The water layer can initiate the shedding of the PVC coating, causing significant decrease in the PVC coating resistance [30] and irreversible deviation of the electrode potential after 16 days of immersion (Fig. 6). After two weeks of immersion, the PVC coating sheds from the GC electrode surface, resulting in the final failure of the Fc–PVC–ZephCl–GC electrode. 4. Conclusion The performance of the PVC-coated GC electrode with ZephCl serving as electrolyte in the PVC coating can significantly improve by adding Fc. This all-solid state ISE can be used as a rapid and reliable sensor for the in-situ detection of TRC in the simulated electrolytically treated BW. Good linear relationship was observed between the electrode potential of the Fc–PVC–ZephCl–GC electrode and the logarithm of TRC within the TRC range of 1 mg/L to 20 mg/L. The response speed and potential recovery of this electrode are satisfactory. No evident interference of the dissolved oxygen and possible co-existing ions was observed on the detection of TRC using this electrode. The stability of this electrode is quite good, and it can provide reliable response to TRC even after six days of immersion in 3.5 wt.% NaCl solution. The failure of the electrode can be ascribed to the dissolution of Fc+ and Zeph+ from the PVC coating, the memory effect of the electrode, and the formation of a water layer between GC electrode and PVC coating. This electrode is a promising sensor for the in-situ detection of TRC in BW to control the electrolysis equipment automatically. Acknowledgments This work is financially supported by the Shandong Provincial Key Innovation Project 2012CX80106. References [1] S. Gollasch, M. David, M. Voigt, E. Dragsund, C. Hewitt, Y. Fukuyo, Critical review of the IMO international convention on the management of ships' ballast water and sediments, Harmful Algae 6 (2007) 585–600. [2] R. Balaji, O. Yaakob, K.K. Koh, A review of developments in ballast water management, Environ. Rev. 22 (2014) 298–310. [3] E. Tsolaki, E. Diamadopoulos, Technologies for ballast water treatment: a review, J. Chem. Technol. Biotechnol. 85 (2010) 19–32. [4] K.G.N. Nanayakkara, Y.M. Zheng, A.K.M.K. Alam, S. Zou, J.P. Chen, Electrochemical disinfection for ballast water management: technology development and risk assessment, Mar. Pollut. Bull. 63 (2011) 119–123. [5] E.C. Kim, K. Shin, K.P. Lee, Development Of Technologies On Test Facility And Procedures For The Land-Based Test As A Type Approval Test At Ballast Water Treatment System, IEEE, 2008 1–7. [6] I. Min, H. Hwang, D. Moon, J. Lee, Implementation of Ballast Water Treatment System Using Electrolysis, IEEE, 2013 1421–1424. [7] I.D. Kim, W.W. Cho, J.Y. Kim, E.C. Nho, G.W. Goh, S.B. Bae, B.N. Kang, Design Of Low-Voltage High-Current Rectifier With High-Efficiency Output Side For Electrolytic Disinfection Of Ballast Water, IEEE, 2010 1652–1657. [8] S. Zhang, X. Chen, D. Yang, W. Gong, Q. Wang, J. Xiao, H. Zhang, Q. Wang, Effects of the chlorination treatment for ballast water, 2nd International Ballast Water Treatment R and D Symposium 2003, pp. 148–157. [9] M.J. Franklin, D.E. Nivens, A.A. Vass, M.W. Mittelman, R.F. Jack, N.J.E. Dowling, D.C. White, Effect of chlorine and chlorine/bromine biocide treatments on the number and activity of biofilm bacteria and on carbon steel corrosion, Corrosion 47 (1991) 128–134. [10] Y.X. Song, K. Dang, H.F. Chi, D.L. Guan, P.H. Yin, Study on effect of metal corrosion of ballast tank caused by ballast water treatment of seawater electrolysis, J. Dalian Marit. Univ. 31 (2005) 45–46.

163

[11] G.Z. Liu, J.M. Wang, J.Q. Zhang, C.N. Cao, Effect of electrolytic treatment of ballast water on corrosion behavior of 316L stainless steel, Acta Metall. Sin. 47 (2011) 1600–1604. [12] D.J. Leggett, N.H. Chen, D.S. Mahadevappa, A flow injection method for analysis of residual chlorine by the DPD procedure, Fresenius Z. Anal. Chem. 315 (1983) 47–50. [13] J.G. March, M. Gual, B.M. Simonet, Determination of residual chlorine in greywater using o-tolidine, Talanta 58 (2002) 995–1001. [14] J.H. Carpenter, C.A. Moore, D.L. Macalady, Errors in determination of residual oxidants in chlorinated seawater, Environ. Sci. Technol. 11 (1977) 992–994. [15] M. Belz, W.J.O. Boyle, K.F. Klein, K.T.V. Grattan, Smart-sensor approach for a fibreoptic-based residual chlorine monitor, Sensors Actuators B 39 (1997) 380–385. [16] S. Saputro, K. Takehara, K. Yoshimura, S. Matsuoka, Narsito, Differential pulse voltammetric determination of free chlorine for water disinfection process, Electroanalysis 22 (2010) 2765–2768. [17] S.Y. Kishioka, T. Kosugi, A. Yamada, Electrochemical determination of a free chlorine residual using cathodic potential-step chronocoulometry, Electroanalysis 17 (2005) 724–726. [18] R. Olivé-Monllau, C.S. Martínez-Cisneros, J. Bartrolí, M. Baeza, F. Céspedes, Integration of a sensitive carbon nanotube composite electrode in a ceramic microanalyzer for the amperometric determination of free chlorine, Sensors Actuators B 151 (2011) 416–422. [19] M. Murata, T.A. Ivandini, M. Shibata, S. Nomura, A. Fujishima, Y. Einaga, Electrochemical detection of free chlorine at highly boron-doped diamond electrodes, J. Electroanal. Chem. 612 (2008) 29–36. [20] K. Senthilkumar, J.M. Zen, Free chlorine detection based on EC mechanism at an electroactive polymelamine-modified electrode, Electrochem. Commun. 46 (2014) 87–90. [21] P. Salazar, M. Martín, F.J. García-García, J.L. González-Mora, A.R. González-Elipe, A novel and improved surfactant-modified Prussian Blue electrode for amperometric detection of free chlorine in water, Sensors Actuators B 213 (2015) 116–123. [22] A.P. Soldatkin, D.V. Gorchkov, C. Martelet, N. Jaffrezic-Renault, New enzyme potentiometric sensor for hypochlorite species detection, Sensors Actuators B 43 (1997) 99–104. [23] A. Sakai, A. Hemmi, H. Hachiya, F. Kobayashi, S. Ito, Y. Asano, T. Imato, Y. Fushinuki, I. Taniguchi, Flow injection analysis for residual chlorine using Pb (II) ion-selective electrode detector, Talanta 45 (1998) 575–581. [24] N. Ishibashi, T. Imato, H. Ohura, S. Yamasaki, Potentiometric flow-injection determination of trace chlorine based on its redox reaction with an iron (III)/iron (II) buffer, Anal. Chim. Acta 214 (1988) 349–357. [25] K.D. Brown, G.A. Parker, Determination of chlorine in concentrated sulfuric acid solution with a residual chlorine electrode, Anal. Chem. 51 (1979) 1332–1333. [26] J.C. Synnot, A.M. Smith, Total residual chlorine by ion-selective electrode-from bench top to continuous monitor, Stud. Environ. Sci. 29 (1986) 777–791. [27] S. Motomizu, T. Yoden, Porous membrane permeation of halogens and its application to the determination of halide ions and residual chlorine by flowinjection analysis, Anal. Chim. Acta 261 (1992) 461–469. [28] J. Bobacka, Conducting polymer-based solid-state ion-selective electrodes, Electroanalysis 18 (2006) 7–18. [29] T. Lindfors, A. Ivaska, Stability of the inner polyaniline solid contact layer in all-solidstate K+-selective electrodes based on plasticized poly (vinyl chloride), Anal. Chem. 76 (2004) 4387–4394. [30] J.P. Veder, R. De Marco, G. Clarke, R. Chester, A. Nelson, K. Prince, E. Pretsch, E. Bakker, Elimination of undesirable water layers in solid-contact polymeric ion-selective electrodes, Anal. Chem. 80 (2008) 6731–6740. [31] M.K.A. El-Rahman, M.R. Rezk, A.M. Mahmoud, M.R. Elghobashy, Design of a stable solid-contact ion-selective electrode based on polyaniline nanoparticles as ion-to-electron transducer for application in process analytical technology as a real-time analyzer, Sensors Actuators B 208 (2015) 14–21. [32] C.Z. Lai, M.A. Fierke, A. Stein, P. Buhlmann, Ion-selective electrodes with threedimensionally ordered macroporous carbon as the solid contact, Anal. Chem. 79 (2007) 4621–4626. [33] E. Jaworska, W. Lewandowski, J. Mieczkowski, K. Maksymiuk, A. Michalska, Simple and disposable potentiometric sensors based on graphene or multi-walled carbon nanotubes–carbon–plastic potentiometric sensors, Analyst 138 (2013) 2363–2371. [34] Z.A. Boeva, K.A. Milakin, M. Pesonen, A.N. Ozerin, V.G. Sergeyev, T. Lindfors, Dispersible composites of exfoliated graphite and polyaniline with improved electrochemical behaviour for solid-state chemical sensor applications, RSC Adv. 4 (2014) 46340–46350. [35] M. Fibbioli, K. Bandyopadhyay, S.G. Liu, L. Echegoyen, O. Enger, F. Diederich, D. Gingery, P. Buhlmann, H. Persson, U.W. Suter, E. Pretsch, Redox-active selfassembled monolayers for solid-contact polymeric membrane ion-selective electrodes, Chem. Mater. 14 (2002) 1721–1729. [36] E. Grygołowicz-Pawlak, K. Wyglądacz, S. Sęk, R. Bilewicz, Z. Brzozka, E. Malinowska, Studies on ferrocene organothiol monolayer as an intermediate phase of potentiometric sensors with gold inner contact, Sensors Actuators B 111 (2005) 310–316. [37] D. Wijesuriya, D.P. Root, Shorter response times and improved selectivity for potentiometric electrodes, Anal. Chim. Acta 236 (1990) 445–448. [38] Q.X. Zha, Electrode Process Kinetics Introduction, 3rd Ed. Science Press, Beijing, 2002. [39] J.S. Do, T.C. Chou, Kinetics of the oxidation of benzyl alcohol by hypochlorite ion in the presence of phase-transfer catalyst, Ind. Eng. Chem. Res. 29 (1990) 1095–1103. [40] R.W. Lee, P.C. Nakagaki, T.C. Bruice, The kinetics for the reaction of hypochlorite with a manganese (III) porphyrin and subsequent epoxidation of alkenes in a homogeneous solution, J. Am. Chem. Soc. 111 (1989) 1368–1372.

164


[41] B. Fu, E. Bakker, J.H. Yun, V.C. Yang, M.E. Meyerhoff, Response mechanism of polymer membrane-based potentiometric polyion sensors, Anal. Chem. 66 (1994) 2250–2259. [42] A. Trojánek, J. Langmaier, Z. Samec, Random nucleation and growth of Pt nanoparticles at the polarised interface between two immiscible electrolyte solutions, J. Electroanal. Chem. 599 (2007) 160–166. [43] J.P. Hurvois, C. Moinet, Reactivity of ferrocenium cations with molecular oxygen in polar organic solvents: decomposition, redox reactions and stabilization, J. Organomet. Chem. 690 (2005) 1829–1839.

[44] X.U. Zou, X.V. Zhen, J.H. Cheong, P. Buhlmann, Calibration-free ionophore-based ion-selective electrodes with a Co (II)/Co (III) redox couple-based solid contact, Anal. Chem. 86 (2014) 8687–8692. [45] U. Oesch, W. Simon, Life time of neutral carrier based ion-selective liquidmembrane electrodes, Anal. Chem. 52 (1980) 692–700.