- Email: [email protected]
Development of a pH sensing membrane electrode based on a new calix[4]arene derivative
Talanta 132 (2015) 669–675
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Development of a pH sensing membrane electrode based on a new calix[4]arene derivative H. Elif Kormalı Ertürün a,n, Ayça Demirel Özel a, Serkan Sayın b, Mustafa Yılmaz b, Esma Kılıç a a b
Department of Chemistry, Faculty of Science, Ankara University, Tandoğan, 06100 Ankara, Turkey Department of Chemistry, Faculty of Science, Selçuk University, Selçuklu, 42075 Konya, Turkey
art ic l e i nf o
a b s t r a c t
Article history: Received 28 August 2014 Received in revised form 10 October 2014 Accepted 14 October 2014 Available online 22 October 2014
A new pH sensing poly(vinyl chloride) (PVC) membrane electrode was developed by using recently synthesized 5,17-bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene as an ionophore. The effects of membrane composition, inner filling solution and conditioning solution on the potential response of the proposed pH sensing membrane electrode were investigated. An optimum membrane composition of 3% ionophore, 67% o-nitrophenyl octyl ether (o-NPOE) as plasticizer, 30% PVC was found. The electrode exhibited a near-Nernstian slope of 58.7 7 1.1 mV pH 1 in the pH range 1.9– 12.7 at 207 1 1C. It showed good selectivity for H þ ions in the presence of some cations and anions and a longer lifetime of at least 12 months when compared with the other PVC membrane pH electrodes reported in the literature. Having a wide working pH range, it was not only applied as a potentiometric indicator electrode in various acid–base titrations, but also successfully employed in different real samples. It has good reproducibility and repeatability with a response time of 6–7 s. Compared to traditional glass pH electrode, it exhibited excellent potentiometric response after being used in fluoride-containing media. & 2014 Elsevier B.V. All rights reserved.
Keywords: Calix[4]arene pH sensing PVC membrane electrode Potentiometry pH measurement Titratable acidity
1. Introduction Accurate and reliable measurement of the pH in chemical, biological, clinical, industrial and environmental samples is always important [1–3]. The pH measurements are generally carried out by use of glass electrodes. They have become very popular due to their selectivity, reliability and dynamic pH range. Although pH-sensitive glass electrode has distinguished response characteristics and has been in use for such a long period, it has certain setbacks such as its high resistance, fragility, its instability in hydrofluoric acid or fluoride solutions, alkaline and acid errors and its unsuitability to serve as a microelectrode for biological applications [4,5]. Accordingly, the studies to develop non-glass pH-sensitive electrodes such as solid contact electrodes (SCE’s) [1,6–13], coated wire electrodes (CWE’s) [13,14], field effect transistors (FET’s) [15,16], nanorod based electrodes [17,18] and polymeric membrane electrodes [19– 33] have gained interest. Among these, PVC membrane electrodes are the preferred ones due to their low electrical resistance and ease of construction [31,32]. In addition, with the membrane supported
n
Corresponding author. Tel.: þ 90 3122126720; fax: þ 90 3122232395. E-mail address: [email protected] (H.E. Kormalı Ertürün).
http://dx.doi.org/10.1016/j.talanta.2014.10.032 0039-9140/& 2014 Elsevier B.V. All rights reserved.
by a tough material, they are solid and not easily fragile. These properties are critical, especially for in-vivo and in-vitro pH measurements. pH-sensitive membrane electrodes based on PVC are also considered for clinical applications and they are currently under investigation [8,10,22,34]. Many authors have focused on the design of new carriers for hydrogen ions to use as ionophores in the construction of pH sensing membrane electrodes. Among these carriers, hexafluorophosphate [1], diazacyclooctadecane [4], decamethylcyclopentasiloxane [6], hexabutyl-tri-amidophosphate [19], hex-3-ene [20], azole [21], phenoxazine derivatives (chromoionophores) [10,27,28], amines [11–13,22–25], pillar[5]arene [29] and calix[4]arenes [30–33] appear to be promising. Calixarenes are very popular as attractive and excellent ionophores due to their host–guest chemistry and ion-recognition properties [35,36]. It is stated in the literature that some specially designed calixarenes could successfully be used for the construction of pH electrodes [30–33]. Therefore, in this work, we focused on the possibility of us ing recently synthesized 5,17-bis(4-benzylpiperidine-1-yl)methyl25,26,27,28-tetrahydroxy calix[4]arene (Fig. 1) as ionophore in the construction of a new pH sensing membrane electrode. It is known that calix[4]arene derivatives bearing amine or amide functions were capable of interacting with anions by hydrogen bonds.
670
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
5,17-Bis-[(4-benzylpiperidine)methyl]-25,26,27,28-tetrahydroxycalix[4]arene was synthesized at the Department of Organic Chemistry of Selçuk University by procedures given in the literature [37]. 2.2. Preparation of the electrode
N
N
OH
OH OH
OH
Fig. 1. The structure of 5,17-bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene.
The pH-sensing membrane electrode was prepared as follows: 0.0128 g of ionophore and 0.2851 g of o-NPOE were dissolved in 5 mL THF. 0.1276 g of PVC was slowly added to this mixture. The homogenous mixture formed was poured onto a glass disc with a diameter of 3.5 cm attached to a glass plate and was kept at room temperature for 24 h for the evaporation of THF. The 0.7 cm diameter disc of the polymer membrane was cut and fixed to the end of a glass tube with a diameter of 0.5 cm and a length of 10 cm. An internal filling solution containing 1.0 10 3 M CaCl2 was put into this glass tube and an AgCl-coated silver wire was placed into it. The prepared electrode was conditioned in deionized water for 12 h before each use. 2.3. Potential measurements
The studies regarding the removal of arsenate and dichromate ions from the aqueous solutions by using the calix[4]arene studied in liquid–liquid extraction indicated that it was protonated at protonswitchable binding sites in acidic conditions [37]. Hence, facilitating binding of anions at low pH is resulted from electrostatic interactions as well as hydrogen bonding via N atoms in upper piperidine rim and –OH groups in lower rim of the calix[4]arene derivative used in this study. It was considered that the calix[4]arene derivative could be used as suitable ionophore for hydrogen ion due to its possible reversible interaction with piperidine nitrogen or –OH groups of this compound. For this purpose, membranes with various combinations were prepared by changing the ratio of ionophore, PVC, plasticizer and lipophilic additive in order to determine the optimum membrane composition. The selectivity against some anions and cations, response time, lifetime, working range and other performance characteristics of the proposed electrode were investigated. The developed pH sensing membrane electrode worked well under laboratory conditions to determine titratable acidity and pH in real samples.
2. Experimental 2.1. Reagents and solutions All chemical substances used were of reagent grade and were used without further purification. High molecular weight poly(vinyl chloride), PVC, o-nitrophenyl octyl ether, o-NPOE, dibutyl phtalate, DBP, bis-(2-ethylhexylsebacate), BEHS, potassium tetrakis(p-chlorophenyl)borate, KTpClPB, sodium tetraphenylborate, NaTPB, cesium tetrakis(3-methylphenyl)borate, CsT(3-Me)PB, and tetrahydrofurane, THF, were obtained from Sigma-Aldrich/Fluka in selectophore purity. Britton–Robinson, BR, buffer solutions were used to investigate the performance of the pH sensing membrane electrode. BR buffer solutions were prepared by mixing appropriate amounts of boric acid, acetic acid and phosphoric acid; and the pH of these solutions were adjusted with the addition of diluted sodium hydroxide or hydrochloric acid solutions to the stock buffer solutions by the use of a combined glass pH-electrode. The working solutions were obtained by dilution of 0.1 M stock solutions of ions with deionized water which was supplied from ELGA Purelab Classic Ultrapure Water System.
All potentiometric measurements were performed at 207 1 1C by using the following cell assembly: Ag/AgCl reference electrode | analyte solution | membrane | 1.0 10 3 M CaCl2 | AgCl | Ag. The potential measurements were made by the use of MettlerToledo SevenMulti pH–Ionmeter with Rondolino Sample Changer. The reference electrode was Ag/AgCl electrode (Mettler-Toledo Inlab Reference, catalogue no: 51343190). The pH measurements were carried out with Mettler-Toledo Inlab Routine Pro pH electrode (catalogue no: 51343055). The test solutions were stirred and the potentials were recorded after the equilibrium potentials had been reached. The potentiometric titrations of hydrochloric, hydrofluoric and phosphoric acids were carried out with automatic titrater, Orion 940 expandable ion analyzer S/N 5816. The proposed electrode was employed in real samples such as vinegar, red wine, beer, fizzy drinks and fruit juices to demonstrate its applicability. All electrochemical measurements were carried out with CHI660D Electrochemical Workstation and the experimental data taken under open circuit potential was recorded with three electrode cell in which Ag/AgCl/KCl, 3 M was the reference electrode. The electrode under study was the working electrode and the auxiliary electrode was platinum wire (BASi MW-1032). The impedance spectra were recorded in solutions of acetate buffer and BR buffer containing sodium chloride within the frequency range of 100 kHz to 0.01 Hz at 0.2 V and the amplitude for the sinusoidal excitation signal was 0.05 mV.
3. Results and discussion 3.1. The determination of the membrane composition In the field of ion-selective electrodes (ISE), the common range of membrane composition was determined as 1–7% ionophore, 28–33% PVC (internal matrix), 60–69% plasticizer (solvent) and 0.03–2% lipophilic anion [38]. Herein, the optimum membrane composition for the proposed pH-sensing membrane electrode was found as 3% ionophore, 67% plasticizer and 30% PVC. The results obtained from the electrodes prepared with different membrane compositions are summarized in Table 1. The SEM images for the membrane of the proposed electrode with and without ionophore were shown in Fig. 2. The surface of both
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
671
Table 1 Effect of the membrane composition on the potentiometric response of new calix[4]arene based pH-sensing membrane electrode (Plasticizer/PVC ¼2.2). Electrode number
1 2 3 4 5 6 7 8 9 a
Membrane composition (wt%) (mT: 425.5 mg) Ionophore
Plasticizer
Additivea
PVC
– 1 2 3 3 3 3 3 3
69.2 (o-NPOE) 69 (o-NPOE) 68.3 (o-NPOE) 67 (o-NPOE) 67 (DBP) 67 (BEHS) 67 (o-NPOE) 67 (o-NPOE) 67 (o-NPOE)
– – – – – – KTpClPB NaTPB CsT(3-Me)PB
30.8 30 29.7 30 30 30 30 30 30
Slope 7 tS/√N (mV pH 1)
Linear pH range
39.5 7 2.2 42.9 7 0.9 53.3 7 1.4 58.7 7 1.1 51.2 7 1.8 24.37 1.3 47.17 2.1 52.2 7 1.7 40.9 7 0.6
3.2–10.7 1.9–10.7 1.9–12.7 1.9–12.7 4.8–10.7 4.8–10.7 3.2–10.7 4.8–12.7 3.2–12.7
N ¼3 N ¼3 N ¼3 N¼ 10 N ¼3 N ¼3 N ¼3 N ¼3 N ¼3
Lipophilic additives were 70% mole ratio relative to the ionophore.
Fig. 2. SEM images of the PVC membranes containing 67 % o-NPOE and 30 % PVC (a) with 3% ionophore (b) without ionophore.
membranes was uniform and there were no defect. It can be seen that white spots in Fig. 2a referred as ionophore, which enhanced the sensitivity of H þ ions of the membrane, was distributed hom ogeneously. Concerning the effect of concentration of the macrocyclic compounds introduced into the membrane upon the selectivity, working range and the slope of the electrode, it is reported that the interaction between the macrocyclic compound and the species to be determined has a notable influence on these features [38]. Therefore, the preliminary studies were carried out by the membranes prepared by keeping the plasticizer/PVC ratio at 2.2 and changing the ratio of ionophore in the range 1–3%. As can be seen in Table 1, the electrode prepared with polymeric membrane containing 1% ionophore (electrode 2) and without ionophore (electrode 1) showed sub-Nernstian slopes in the linear working range of pH 1.9–10.7 and pH 3.2–10.7, respectively. It is stated in the literature that the complex stoichiometry between the macrocyclic compound and the analyte ion is of great importance regarding the selectivity and the slope of the electrode [38,39]. Therefore, the percentage of the ionophore in the membrane was increased up to 2 and 3% in order to investigate the effect of the ionophore concentration upon the performance. Not only the slopes of electrode 3 (53.3 71.4 mV pH 1) and electrode 4 (58.7 71.1 mV pH 1) were improved but also the linear working ranges were observed to be better than those containing 1% ionophore. We concluded that increasing the ionophore ratio upto 3% provided better response characteristics and the studies related to optimization of the membrane composition were continued by keeping the ionophore ratio at 3%. Due to the fact that the plasticizers change the dielectric constant of the membrane phase, the type of the plasticizer has
a remarkable effect upon the selectivity and the performance characteristics of the ISE membranes [40–42]. Therefore, DBP, BEHS and o-NPOE were used to investigate the effect of the plasticizer. It is clear in Table 1 that the electrode containing o-NPOE (electrode 4) exhibited higher slope and wider linear working range than that of the electrodes prepared with DBP (electrode 5) and BEHS (electrode 6). This led us to conclude that o-NPOE was the best membrane plasticizer for the proposed pH-sensing membrane electrode. That was why all the studies to determine the performance of the electrode and all the analytical applications relating to it were performed by the use of this electrode. It is well-known that the addition of a lipophilic anion to neutral carrier membranes improved almost all of the electrode characteristics, since it stabilizes the membrane operating conditions, decreases the membrane resistance and significantly reduces the response times [39,43,44]. In our study, the electrodes were prepared by using KTpClPB (electrode 7), NaTPB (electrode 8), and CsTp (3-Me)PB (electrode 9) as lipophilic additives. These electrodes showed weaker H þ selectivity with sub-Nernstian slopes and narrower pH ranges. This is not in accordance with the literature, since it is claimed that the addition of lipophilic sites to the membrane phase improves the performance of the electrodes. It may be attributed to the fact that sodium, potassium and cesium ions which form ion-pairs with tetraphenylborate and tetrakis(p-chlorophenyl)borate, settle in the cavity of the calix[4] arene and decrease the number of active sites in the membrane. The adverse effect in our case can also be the result of the nature of the ionophore. Taking into account both the slopes and the linear pH range of the electrodes, the electrode prepared with the membrane containing 3% ionophore, 67% o-NPOE and 30% PVC demonstrated the best Nernstian slope of 58.7 71.1 mV pH 1 in the widest pH range
672
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
1.9–12.7 (electrode 4). Hence, further studies were performed with the electrode 4.
500 (a)
400
y = -58.888x + 566.6 R² = 0.9999
3.2. The effect of conditioning and inner filling solution
(b)
300
y = -59.842x + 501.53 R² = 0.994
E, mV
200 100 0 -100 -200 -300
2
4
6
8
10
12
14
pH Fig. 3. Typical calibration graphs of the proposed pH sensing membrane electrode (electrode 4) plotted by the data obtained from (a) pH-ion meter and (b) potentialtime graph.
400 1.96
2.99
300
3.97
4.89 5.92
200
E, mV
The effect of the conditioning upon the performance of the electrode was investigated by the use of two electrodes with the same membrane composition and 1.0 10 3 M CaCl2 as inner filling solution. The electrodes were kept in 1.0 10 3 M HCl and in deionized water for 12 h, separately. The results obtained revealed that the performance and the lifetime of the electrode conditioned in deionized water were higher than that of the electrode conditioned in acid solution. Therefore, the electrodes were kept in deionized water prior to each experiment. The composition and the concentration of the inner filling solution were found to have a significant effect on the potentiometric response of the PVC membrane electrodes by means of changing its selectivity and working range [45]. Either the chloride salts of inert electrolyte or the chloride salt of analyte ion with lower concentration were commonly used as internal solutions [46,47]. For this purpose, the electrode prepared with the membrane containing 3% ionophore, 67% o-NPOE and 30% PVC was filled with 1.0 10 3 M HCl and 1.0 10 3 M CaCl2. From the view point of linear working range, it was determined that the most appropriate inner filling solution was 1.0 10 3 M CaCl2. This application also confirmed that inner filling solutions prepared with inert salts have great effect on the working range of the electrodes [47].
100
6.61 7.84 8.66
0
3.3. Working range and the slope of the electrode
-100
In order to investigate the working range and the slope of the electrode under study, two different types of calibration graphs were plotted (Fig. 3). The potentials measured were plotted against the pH values to obtain the first calibration curve (Fig. 3a) by using series of buffer solutions in the range of pH 1.9–12.9. The second calibration curve (Fig. 3b) was performed by recording the time elapsed to reach a stable potential value after the addition of diluted NaOH to access the appropriate pH (see Fig. 4). The proposed electrode was found much more suitable than many of the other alternatives given in the literature as regard to the linear working range and slope [1,6–14,16,18–20,22– 25,30,32,33]. The comparison of response characteristics of some different type of pH sensing electrodes together with the proposed electrode are given in Table 2. In this study, the impedance measurements were evaluated in two different buffer solutions containing sodium chloride. Fig. 5 (A) displays the impedance plots of the proposed pH sensing membrane and the same membrane without ionophore immersed into acetic acid/acetate buffer-NaCl solution (pH: 4). This type of spectra shows the behaviour of the membranes in the electrodes. As it can be seen, at high frequencies, the signal is mainly dominated by the electrical resistance and the geometric capacitance in parallel with the membrane which showed well defined semi-cyclic impedances [1,7]. The electrical resistance for the ionophore free membrane (electrode 1) was much lower than the resistance for the ionophore-containing membrane (electrode 4). On the contrary, it was stated in literature that the addition of neutral carriers such as alkyldibenzylamines had no effect on the electrical resistance of the membrane electrodes [24]. This behaviour may be attributed to the introduction of a huge neutral macrocyclic compound as calix[4]arene derivative to the membrane. On the other hand, there is a diffusional component ( 451 line) at low frequencies which clearly indicates the reversible diffusion of the hydrogen ion from the solution into the proposed electrode membrane [7].
-200
9.80 11.30 12.22
-300 0
100
200
300
400
500
600
700
800
Time, sec Fig. 4. Dynamic response of the proposed electrode for step changes in pH (electrode 4).
The impedance behavior of the proposed electrode in different pH buffered solutions (pH 3, pH 5 and pH 7) was presented in Fig. 5(B). The electrode displayed nearly the same semi-cyclic impedances at high frequencies. However, the bulk resistance increased with increasing pH values. This phenomenon was related to the transfer of hydrogen ion from solution to the calix [4]arene derivative in polymer layer. Hence, it is in accordance with the literature proposing that the energy of transfer of the hydrogen ion across the polymer membrane increased with increasing pH values [12,13]. 3.4. Response time and lifetime of the electrode The response time of an ion-selective electrode is an important factor for any analytical application. Therefore, the response time required for the electrode to reach 95% of equilibrium mV values after successful immersion in a series of solution each having a 10fold difference in hydrogen ion concentration was measured. The static response time thus obtained was 6–7 s over the entire pH range (1.9–12.7) and no change was observed in 10 min. This was also confirmed from the potential-time graph in Fig. 4. The response times are comparable to, or even shorter than those reported in literature, that were found to give linear response in the same or narrower working ranges [6–12,30–34]. The lifetime of the electrode was determined by reading its potentials and plotting the calibration curves for a period of 12
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
673
Table 2 Comparison of the proposed electrode with some pH sensing electrodes. Electrode
Sensing material
Linear range (pH)
Sensitivity (mV pH 1)
Response time (t95)
References
RSE SCE CWE FET MOE PME PME PME
Quinhydrone Decamethylcyclopentasiloxane N,N-dioctadecylmethylamine Carbon-nanotube thin film MoO3 nanorod Calix[4]-aza-crown derivative Tri-n-dodecylamine Calix[4]arene derivative
2.0–9.5 1.9–9.8 4.0–11.0 3.0–13.0 1.0–5.0 2.5–11.5 4.0–10.0 1.9–12.7
57.5 70.2 57.6 70.2 50.2 7 0.4 50.9 547 2 59.2 547 2 58.7 7 1.1
o 10 s o 15 s o5 s
[1] [6] [14] [16] [18] [30] [22] This work
o 60 s o 10 s o6 s 6–7 s
RSE redox system based electrode; SCE solid contact electrode; CWE coated wire electrode; FET field effect transistor; MOE metal oxide electrode; PME polymeric membrane electrode.
0.8
a b
1.0 0.8
-Z'', MΩ
-Z′′, MΩ
0.6
0.4
0.2
0.6 0.4 0.2 0.0
0.0 0.0
c d e
0.5
1.0
1.5
2.0
2.5
Z′, MΩ
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Z', MΩ
Fig. 5. Typical impedance plots of the membranes containing o-NPOE and PVC (A) in acetic acid/acetate buffer–NaCl solution (pH: 4): (a) without ionophore, electrode 1, (b) with 3% ionophore, electrode 4; (B) in BR buffer-NaCl solutions at: (c) pH: 3, (d) pH: 5, (e) pH: 7 for electrode 4.
months. The slope of the electrode was observed to show no significant change even after this period. This shows that the lifetime of the electrode prepared by the use of 5,17-bis(4benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4] arene was longer than the PVC membrane and other non-glass pH electrodes reported in literature with good repeatability and reproducibility [12,20,22,30–32]. 3.5. The selectivity of the electrode It is reported in literature that the working range of macrocyclic compounds-based electrodes was adversely affected by the anions and the cations. It is known that cations decrease the lower detection limit (at high pH) and anions decrease the upper detection limit (at low pH) of the pH electrodes [23,33]. In short, many ionic impurities are expected to narrow the working range of the electrode [13,43]. The interferences of cations such as Na þ , K þ , Li þ , NH4þ , Mg2 þ , Ca2 þ (0.1 M) and anions such as bromide, iodide, chloride, nitrate, thiocyanate, perchlorate (1.0 10 3 M) were evaluated by the use of the fixed interference method (FIM). The effect of the above mentioned anions on the pH response of the electrode was demonstrated in Fig. 6a. Tentatively notwithstanding, in the presence of higher anion concentrations at low pH values, anions form ion pair with hydrogen ions which are bonded to calix [4]arene thereby making the ionophore irresponsible, for the protons, to penetrate into and the system behaves insensitive to hydrogen ions. In this case, the potential is independent of the pH and is a function of the concentration of anions in the solution [31,32]. As it can be seen in Fig. 6a, the anions except from universal impurity chloride affected the working range of the proposed pH sensing membrane electrode. The following sequence of anion interference was observed: ClO4 4SCN 4I 4NO3 4Br .
Cation interference was also investigated in the region of high pH, where the hydrogen ion concentration is very low, but the above mentioned cation concentration is high, that causes alkaline error in case of some glass pH electrodes. The potentiometric response of the electrode in the presence of these cations indicated that the working range and the Nernstian slope were not changed (Fig. 6b). These results indicated that the corresponding cations would not cause any significant interference unless they present in concentrations 40.1 M. These results are comparable to and even better than some of the data given for other PVC-based membrane electrodes in the literature [6,14,31–33]. Therefore, this electrode can conveniently be used for pH measurements in the media containing these ions.
3.6. The analytical applications of the electrode The electrode was used in the titrations of hydrochloric, hydrofluoric and phosphoric acids with standardized sodium hydroxide. The titration curves were plotted in order to elucidate whether the prepared pH sensing membrane electrode can be employed as an indicator electrode in acid–base titrations. The same titrations were also carried out by the use of a commercial combined glass pH electrode. The expected end-points were observed identical for the proposed electrode and combined glass pH electrode. An example of each titration curves is given as supplementary material (Fig. 7). Furthermore, no erosion effect of hydrofluoric acid was observed on the prepared membrane. After a measurement with this electrode in hydrofluoric acid, the working range and the Nernstian slope remained the same. It can be concluded that the electrode prepared could be successfully employed in acid–base titrations and in fluoride containing media.
674
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
500
500 + H Br I Cl
300
E, mV
300
NO3 ClO4 SCN
200 100 0
+ H + Na + K
400
+ NH4 + Li 2+ Mg 2+ Ca
200
E, mV
400
100 0 -100
-100
-200
-200
-300 0
2
4
6
8
10
12
14
0
2
4
pH
6
8
10
12
14
pH
Fig. 6. Potentiometric response of the proposed electrode in the presence of (a) anions (C ¼1.0 10 3 M), (b) cations (C¼ 0.1 M).
Table 3 The titratable total acidity of beverages found with the proposed electrode and commercial combined glass pH electrode (95% CL, N ¼5, tcrit ¼ 2.78, Fcrit ¼ 6.39).a Beverages
Glass pH electrode
Proposed electrode
texp
Fexp
Orange juice Apple juice Fizzy drink Beer Red wine Vinegar
13.3 7 0.6 9.8 7 0.3 0.6 7 0.0 0.6 7 0.0 8.4 7 0.2 85.3 7 4.3
13.3 7 0.6 9.8 7 0.1 0.7 7 0.0 0.6 7 0.0 8.17 0.2 83.6 7 1.9
0.1 0.4 2.3 0.1 2.1 0.9
1.0 0.2 0.2 1.0 0.4 0.2
a
The results were rounded as one digit beyond the decimal point.
The feasibility of the proposed electrode for the determination of the titratable acidity of some commercial beverages was also investigated [48]. Packaged of beverages were obtained from a local market (Ankara, Turkey). The analytical results obtained by this pH electrode were compared with those obtained by commercial combined glass pH electrode (Table 3). From Table 3, the critical value of t is 2.78 for the 95% confidence level and 4 degree of freedom. Since t4texp, no significant differences between the results obtained with the proposed pH sensing PVC membrane electrode and the commercial combined glass pH electrode was detected. Furthermore, since critical value of F(6.39) 4 Fexp, we concluded that the results obtained from the proposed electrode appear to give better precision than those from combined glass pH electrode at the 95% confidence level. It can be suggested that the developed calix[4]arene based PVC membrane electrode could be an effective tool for potentiometric determination of the titratable acidity in real samples.
4. Conclusions 5,17-Bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene can be successfully used as an ionophore in pHsensing membrane electrode. The proposed electrode can be a good alternative for a glass electrode. It showed good selectivity for H þ ions in the presence of some cations and anions and a longer lifetime of at least 12 months. This electrode was also found much more suitable when compared with the other PVC membrane pH electrodes reported in the literature as regard to the linear working range and slope. Due to the fact that the electrode had a wide working pH range, it can be not only employed as an indicator electrode in acid–base titrations, but also can be used for the titratable acidity in different real samples. It was also
successfully utilized in fluoride-containing media. The advantages of this electrode are its simplicity of preparation, low-cost, fast response time and long lifetime. Furhermore, it can be possible to prepare a PVC micro-pH electrode using the new calix[4]arene derivative to perform in-vivo pH measurements. We predicted that the proposed electrode could also be used as an anionselective electrode for some anions due to the ability of protonated calix[4]arene binding anions reversibly to the polymer membrane by hydrogen bonds [49].
Acknowledgements We gratefully acknowledge the financial support of Ankara University Research Fund (Grant no: 10B4240003).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.10.032. References [1] J. Ping, Y. Wang, J. Wu, Y. Ying, F. Ji, Microchim. Acta 176 (2012) 229–234. [2] I.M. Illyas, H.H. Mohd, S.S. Intan, A.G Sulaiman, Int. J. Electrochem. Sci. 7 (2012) 12045–12053. [3] X. Ge, Y. Kostov, L. Tolosa, G. Rao, Anal. Chim. Acta 734 (2012) 79–87. [4] R. Ansari, M. Arvand, L. Heydari, J. Chem. Sci. 126 (2014) 41–48. [5] L.B. Peng, L.Y. Heng, S.A. Hasbullah, M. Ahmad, J. Anal. Chem. 62 (2007) 884–888. [6] B. Kim, J. Shim, K.C. Chung, Anal. Lett. 44 (2011) 2138–2149. [7] G.A. Crespo, D. Gugsa, S. Macho, F.X. Rius, Anal. Bioanal. Chem. 395 (2009) 2371–2376. [8] A. Michalska, A. Hulanicki, A. Lewenstam, Analyst 119 (1994) 2417–2420. [9] R.C. Faria, L.O.S. Bulhões, Anal. Chim. Acta 377 (1998) 21–27. [10] N. Zine, J. Bausells, A. Ivorra, J. Aguiló, M. Zabala, F. Teixidor, C. Masalles, C. Viñas, A. Errachid, Sens. Actuators, B 91 (2003) 76–82. [11] W.S. Han, K.C. Chung, M.H. Kim, H.B. Ko, Y.H. Lee, T.K. Hong, Anal. Sci. 20 (2004) 1419–1422. [12] W.S. Han, M.Y. Park, K.C. Chung, K.C. Cho, T.K. Hong, Electroanal 13 (2001) 955–959. [13] W.S. Han, M.Y. Park, K.C. Chung, K.C. Cho, T.K. Hong, Talanta 54 (2001) 153–159. [14] P.W. Alexander, T. Dimitrakopoulos, D.B. Hibbert, Electroanalysis 9 (1997) 1331–1336. [15] J.H. Ahn, J.Y. Kim, M.L. Seol, D.J. Baek, Z. Guo, C.H. Kim, S.J. Choi, Y.K. Choi, Appl. Phys. Lett. 102 (2013) 1–5. [16] Y.S. Chien, W.L. Tsai, I.C. Lee, J.C. Chou, H.C. Cheng, IEEE Electron Device Lett. 33 (2012) 1622–1624. [17] M. Kashif, M.E. Ali, S.M.U. Ali, U. Hashim, S.B.A. Hamid, Nanoscale Res. Lett. 8 (2013) 68–77. [18] G.S. Zakharova, N.V. Podval’naya, J. Anal. Chem. 68 (2013) 50–56.
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
[19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]
V.M. Lutov, K.N. Mikhelson, Sens. Actuators, B 18 (1994) 400–403. M. Arvand, K. Ghaiuri, Talanta 79 (2009) 863–870. J. Chojnacki, J.F. Biernat, J. Electroanal. Chem. 277 (1990) 159–164. K. Nurminen, L. Outinen-Y, S. Narkilahti, J. Lekkala, Procedia Eng. 5 (2010) 544–547. D.H. Cho, K.C. Chung, S.S. Jeong, M.Y. Park, Talanta 51 (2000) 761–767. D.H. Cho, K.C. Chung, M.Y. Park, Talanta 47 (1998) 815–821. K.I. Joung, H.J. Yoon, H. Nam, K.J. Paeng, Microchem. J. 68 (2001) 115–120. G.A. Crespo, M.G. Afshar, E. Bakker, Anal. Chem. 84 (2012) 10165–10169. E. Lindner, V.V. Cosofret, R.P. Kusy, R.P. Buck, Talanta 40 (1993) 957–967. J. Langmaier, E. Lindner, Anal. Chim. Acta 543 (2005) 156–176. R.R. Kothur, J. Hall, B.A. Patel, C.L. Leong, M.G. Boutelle, P.J. Cragg, Chem. Commun. 50 (2014) 852–854. X.J. Liu, B. Peng, F. Liu, Y. Qin, Sens. Actuators, B 125 (2007) 656–663. D. Kuruoğlu, E. Canel, S. Memon, M. Yılmaz, E. Kılıc, Anal. Sci. 19 (2003) 217–221. A. Demirel, A. Doğan, E. Canel, S. Memon, M. Yılmaz, E. Kılıc, Talanta 62 (2004) 123–129. V.V. Egorov, Y.V. Sin’kevich, Talanta 48 (1999) 23–28. C. Espadas-Torre, M.E. Meyerhoff, Anal. Chem. 67 (1995) 3108–3114. E.M. Collins, M.A. McKervey, E. Madigan, M.B. Moran, M. Owens, G. Ferguson, S.J. Harris, J. Chem. Soc., Perkin Trans. 1 (1991) 3137–3142.
675
[36] C.D. Gutsche, K.C. Nam, J. Am. Chem. Soc. 110 (1988) 6153–6162. [37] S. Sayin, F. Ozcan, M. Yilmaz, A. Tor, S. Memon, Y. Cengeloglu, Clean—Soil, Air, Water 38 (2010) 639–648. [38] Y.A. Zolotov, Macrocyclic Compounds in Analytical Chemistry, John Wiley and Sons Ltd., USA, 1997. [39] U. Schaller, E. Bakker, E. Spichiger, E. Pretsch, Anal. Chem. 66 (1994) 391–398. [40] Y.W. Choi, N. Minoura, S.H. Moon, Talanta 66 (2005) 1254–1263. [41] Y.W. Choi, S.H. Moon, Environ. Monit. Assess. 70 (2001) 167–180. [42] M.M.G. Antonisse, D.N. Reinhoudt, Electroanalysis 11 (1999) 1035–1048. [43] S. Amemiya, P. Bühlmann, Y. Umezawa, Anal. Chem. 70 (1998) 445–454. [44] R. Eugster, P.M. Gehrig, W.E. Morf, U.E. Spichiger, W. Simon, Anal. Chem. 63 (1991) 2285–2289. [45] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81–84. [46] J Bobacka, V. Väänänen, A. Lewenstam, A. Ivaska, Talanta 63 (2004) 135–138. [47] E. Bakker, P. Buhlmann, E. Pretsch, Electroanalysis 11 (1999) 915–933. [48] Anonim, TS 1125 ISO 750, Meyve ve Sebze Ürünleri Titre Edilebilir Asitlik Tayini, Turkish Standards Institution, Ankara, 2002 (Translated from ISO 750:1998(EN) Fruit and Vegetable Products—Determination of Titratable Acidity). [49] S. Erden, A. Demirel, S. Memon, M. Yılmaz, E. Canel, E. Kılıç, Sens. Actuators, B 113 (1) (2006) 290–296.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Development of a pH sensing membrane electrode based on a new calix[4]arene derivative H. Elif Kormalı Ertürün a,n, Ayça Demirel Özel a, Serkan Sayın b, Mustafa Yılmaz b, Esma Kılıç a a b
Department of Chemistry, Faculty of Science, Ankara University, Tandoğan, 06100 Ankara, Turkey Department of Chemistry, Faculty of Science, Selçuk University, Selçuklu, 42075 Konya, Turkey
art ic l e i nf o
a b s t r a c t
Article history: Received 28 August 2014 Received in revised form 10 October 2014 Accepted 14 October 2014 Available online 22 October 2014
A new pH sensing poly(vinyl chloride) (PVC) membrane electrode was developed by using recently synthesized 5,17-bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene as an ionophore. The effects of membrane composition, inner filling solution and conditioning solution on the potential response of the proposed pH sensing membrane electrode were investigated. An optimum membrane composition of 3% ionophore, 67% o-nitrophenyl octyl ether (o-NPOE) as plasticizer, 30% PVC was found. The electrode exhibited a near-Nernstian slope of 58.7 7 1.1 mV pH 1 in the pH range 1.9– 12.7 at 207 1 1C. It showed good selectivity for H þ ions in the presence of some cations and anions and a longer lifetime of at least 12 months when compared with the other PVC membrane pH electrodes reported in the literature. Having a wide working pH range, it was not only applied as a potentiometric indicator electrode in various acid–base titrations, but also successfully employed in different real samples. It has good reproducibility and repeatability with a response time of 6–7 s. Compared to traditional glass pH electrode, it exhibited excellent potentiometric response after being used in fluoride-containing media. & 2014 Elsevier B.V. All rights reserved.
Keywords: Calix[4]arene pH sensing PVC membrane electrode Potentiometry pH measurement Titratable acidity
1. Introduction Accurate and reliable measurement of the pH in chemical, biological, clinical, industrial and environmental samples is always important [1–3]. The pH measurements are generally carried out by use of glass electrodes. They have become very popular due to their selectivity, reliability and dynamic pH range. Although pH-sensitive glass electrode has distinguished response characteristics and has been in use for such a long period, it has certain setbacks such as its high resistance, fragility, its instability in hydrofluoric acid or fluoride solutions, alkaline and acid errors and its unsuitability to serve as a microelectrode for biological applications [4,5]. Accordingly, the studies to develop non-glass pH-sensitive electrodes such as solid contact electrodes (SCE’s) [1,6–13], coated wire electrodes (CWE’s) [13,14], field effect transistors (FET’s) [15,16], nanorod based electrodes [17,18] and polymeric membrane electrodes [19– 33] have gained interest. Among these, PVC membrane electrodes are the preferred ones due to their low electrical resistance and ease of construction [31,32]. In addition, with the membrane supported
n
Corresponding author. Tel.: þ 90 3122126720; fax: þ 90 3122232395. E-mail address: [email protected] (H.E. Kormalı Ertürün).
http://dx.doi.org/10.1016/j.talanta.2014.10.032 0039-9140/& 2014 Elsevier B.V. All rights reserved.
by a tough material, they are solid and not easily fragile. These properties are critical, especially for in-vivo and in-vitro pH measurements. pH-sensitive membrane electrodes based on PVC are also considered for clinical applications and they are currently under investigation [8,10,22,34]. Many authors have focused on the design of new carriers for hydrogen ions to use as ionophores in the construction of pH sensing membrane electrodes. Among these carriers, hexafluorophosphate [1], diazacyclooctadecane [4], decamethylcyclopentasiloxane [6], hexabutyl-tri-amidophosphate [19], hex-3-ene [20], azole [21], phenoxazine derivatives (chromoionophores) [10,27,28], amines [11–13,22–25], pillar[5]arene [29] and calix[4]arenes [30–33] appear to be promising. Calixarenes are very popular as attractive and excellent ionophores due to their host–guest chemistry and ion-recognition properties [35,36]. It is stated in the literature that some specially designed calixarenes could successfully be used for the construction of pH electrodes [30–33]. Therefore, in this work, we focused on the possibility of us ing recently synthesized 5,17-bis(4-benzylpiperidine-1-yl)methyl25,26,27,28-tetrahydroxy calix[4]arene (Fig. 1) as ionophore in the construction of a new pH sensing membrane electrode. It is known that calix[4]arene derivatives bearing amine or amide functions were capable of interacting with anions by hydrogen bonds.
670
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
5,17-Bis-[(4-benzylpiperidine)methyl]-25,26,27,28-tetrahydroxycalix[4]arene was synthesized at the Department of Organic Chemistry of Selçuk University by procedures given in the literature [37]. 2.2. Preparation of the electrode
N
N
OH
OH OH
OH
Fig. 1. The structure of 5,17-bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene.
The pH-sensing membrane electrode was prepared as follows: 0.0128 g of ionophore and 0.2851 g of o-NPOE were dissolved in 5 mL THF. 0.1276 g of PVC was slowly added to this mixture. The homogenous mixture formed was poured onto a glass disc with a diameter of 3.5 cm attached to a glass plate and was kept at room temperature for 24 h for the evaporation of THF. The 0.7 cm diameter disc of the polymer membrane was cut and fixed to the end of a glass tube with a diameter of 0.5 cm and a length of 10 cm. An internal filling solution containing 1.0 10 3 M CaCl2 was put into this glass tube and an AgCl-coated silver wire was placed into it. The prepared electrode was conditioned in deionized water for 12 h before each use. 2.3. Potential measurements
The studies regarding the removal of arsenate and dichromate ions from the aqueous solutions by using the calix[4]arene studied in liquid–liquid extraction indicated that it was protonated at protonswitchable binding sites in acidic conditions [37]. Hence, facilitating binding of anions at low pH is resulted from electrostatic interactions as well as hydrogen bonding via N atoms in upper piperidine rim and –OH groups in lower rim of the calix[4]arene derivative used in this study. It was considered that the calix[4]arene derivative could be used as suitable ionophore for hydrogen ion due to its possible reversible interaction with piperidine nitrogen or –OH groups of this compound. For this purpose, membranes with various combinations were prepared by changing the ratio of ionophore, PVC, plasticizer and lipophilic additive in order to determine the optimum membrane composition. The selectivity against some anions and cations, response time, lifetime, working range and other performance characteristics of the proposed electrode were investigated. The developed pH sensing membrane electrode worked well under laboratory conditions to determine titratable acidity and pH in real samples.
2. Experimental 2.1. Reagents and solutions All chemical substances used were of reagent grade and were used without further purification. High molecular weight poly(vinyl chloride), PVC, o-nitrophenyl octyl ether, o-NPOE, dibutyl phtalate, DBP, bis-(2-ethylhexylsebacate), BEHS, potassium tetrakis(p-chlorophenyl)borate, KTpClPB, sodium tetraphenylborate, NaTPB, cesium tetrakis(3-methylphenyl)borate, CsT(3-Me)PB, and tetrahydrofurane, THF, were obtained from Sigma-Aldrich/Fluka in selectophore purity. Britton–Robinson, BR, buffer solutions were used to investigate the performance of the pH sensing membrane electrode. BR buffer solutions were prepared by mixing appropriate amounts of boric acid, acetic acid and phosphoric acid; and the pH of these solutions were adjusted with the addition of diluted sodium hydroxide or hydrochloric acid solutions to the stock buffer solutions by the use of a combined glass pH-electrode. The working solutions were obtained by dilution of 0.1 M stock solutions of ions with deionized water which was supplied from ELGA Purelab Classic Ultrapure Water System.
All potentiometric measurements were performed at 207 1 1C by using the following cell assembly: Ag/AgCl reference electrode | analyte solution | membrane | 1.0 10 3 M CaCl2 | AgCl | Ag. The potential measurements were made by the use of MettlerToledo SevenMulti pH–Ionmeter with Rondolino Sample Changer. The reference electrode was Ag/AgCl electrode (Mettler-Toledo Inlab Reference, catalogue no: 51343190). The pH measurements were carried out with Mettler-Toledo Inlab Routine Pro pH electrode (catalogue no: 51343055). The test solutions were stirred and the potentials were recorded after the equilibrium potentials had been reached. The potentiometric titrations of hydrochloric, hydrofluoric and phosphoric acids were carried out with automatic titrater, Orion 940 expandable ion analyzer S/N 5816. The proposed electrode was employed in real samples such as vinegar, red wine, beer, fizzy drinks and fruit juices to demonstrate its applicability. All electrochemical measurements were carried out with CHI660D Electrochemical Workstation and the experimental data taken under open circuit potential was recorded with three electrode cell in which Ag/AgCl/KCl, 3 M was the reference electrode. The electrode under study was the working electrode and the auxiliary electrode was platinum wire (BASi MW-1032). The impedance spectra were recorded in solutions of acetate buffer and BR buffer containing sodium chloride within the frequency range of 100 kHz to 0.01 Hz at 0.2 V and the amplitude for the sinusoidal excitation signal was 0.05 mV.
3. Results and discussion 3.1. The determination of the membrane composition In the field of ion-selective electrodes (ISE), the common range of membrane composition was determined as 1–7% ionophore, 28–33% PVC (internal matrix), 60–69% plasticizer (solvent) and 0.03–2% lipophilic anion [38]. Herein, the optimum membrane composition for the proposed pH-sensing membrane electrode was found as 3% ionophore, 67% plasticizer and 30% PVC. The results obtained from the electrodes prepared with different membrane compositions are summarized in Table 1. The SEM images for the membrane of the proposed electrode with and without ionophore were shown in Fig. 2. The surface of both
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
671
Table 1 Effect of the membrane composition on the potentiometric response of new calix[4]arene based pH-sensing membrane electrode (Plasticizer/PVC ¼2.2). Electrode number
1 2 3 4 5 6 7 8 9 a
Membrane composition (wt%) (mT: 425.5 mg) Ionophore
Plasticizer
Additivea
PVC
– 1 2 3 3 3 3 3 3
69.2 (o-NPOE) 69 (o-NPOE) 68.3 (o-NPOE) 67 (o-NPOE) 67 (DBP) 67 (BEHS) 67 (o-NPOE) 67 (o-NPOE) 67 (o-NPOE)
– – – – – – KTpClPB NaTPB CsT(3-Me)PB
30.8 30 29.7 30 30 30 30 30 30
Slope 7 tS/√N (mV pH 1)
Linear pH range
39.5 7 2.2 42.9 7 0.9 53.3 7 1.4 58.7 7 1.1 51.2 7 1.8 24.37 1.3 47.17 2.1 52.2 7 1.7 40.9 7 0.6
3.2–10.7 1.9–10.7 1.9–12.7 1.9–12.7 4.8–10.7 4.8–10.7 3.2–10.7 4.8–12.7 3.2–12.7
N ¼3 N ¼3 N ¼3 N¼ 10 N ¼3 N ¼3 N ¼3 N ¼3 N ¼3
Lipophilic additives were 70% mole ratio relative to the ionophore.
Fig. 2. SEM images of the PVC membranes containing 67 % o-NPOE and 30 % PVC (a) with 3% ionophore (b) without ionophore.
membranes was uniform and there were no defect. It can be seen that white spots in Fig. 2a referred as ionophore, which enhanced the sensitivity of H þ ions of the membrane, was distributed hom ogeneously. Concerning the effect of concentration of the macrocyclic compounds introduced into the membrane upon the selectivity, working range and the slope of the electrode, it is reported that the interaction between the macrocyclic compound and the species to be determined has a notable influence on these features [38]. Therefore, the preliminary studies were carried out by the membranes prepared by keeping the plasticizer/PVC ratio at 2.2 and changing the ratio of ionophore in the range 1–3%. As can be seen in Table 1, the electrode prepared with polymeric membrane containing 1% ionophore (electrode 2) and without ionophore (electrode 1) showed sub-Nernstian slopes in the linear working range of pH 1.9–10.7 and pH 3.2–10.7, respectively. It is stated in the literature that the complex stoichiometry between the macrocyclic compound and the analyte ion is of great importance regarding the selectivity and the slope of the electrode [38,39]. Therefore, the percentage of the ionophore in the membrane was increased up to 2 and 3% in order to investigate the effect of the ionophore concentration upon the performance. Not only the slopes of electrode 3 (53.3 71.4 mV pH 1) and electrode 4 (58.7 71.1 mV pH 1) were improved but also the linear working ranges were observed to be better than those containing 1% ionophore. We concluded that increasing the ionophore ratio upto 3% provided better response characteristics and the studies related to optimization of the membrane composition were continued by keeping the ionophore ratio at 3%. Due to the fact that the plasticizers change the dielectric constant of the membrane phase, the type of the plasticizer has
a remarkable effect upon the selectivity and the performance characteristics of the ISE membranes [40–42]. Therefore, DBP, BEHS and o-NPOE were used to investigate the effect of the plasticizer. It is clear in Table 1 that the electrode containing o-NPOE (electrode 4) exhibited higher slope and wider linear working range than that of the electrodes prepared with DBP (electrode 5) and BEHS (electrode 6). This led us to conclude that o-NPOE was the best membrane plasticizer for the proposed pH-sensing membrane electrode. That was why all the studies to determine the performance of the electrode and all the analytical applications relating to it were performed by the use of this electrode. It is well-known that the addition of a lipophilic anion to neutral carrier membranes improved almost all of the electrode characteristics, since it stabilizes the membrane operating conditions, decreases the membrane resistance and significantly reduces the response times [39,43,44]. In our study, the electrodes were prepared by using KTpClPB (electrode 7), NaTPB (electrode 8), and CsTp (3-Me)PB (electrode 9) as lipophilic additives. These electrodes showed weaker H þ selectivity with sub-Nernstian slopes and narrower pH ranges. This is not in accordance with the literature, since it is claimed that the addition of lipophilic sites to the membrane phase improves the performance of the electrodes. It may be attributed to the fact that sodium, potassium and cesium ions which form ion-pairs with tetraphenylborate and tetrakis(p-chlorophenyl)borate, settle in the cavity of the calix[4] arene and decrease the number of active sites in the membrane. The adverse effect in our case can also be the result of the nature of the ionophore. Taking into account both the slopes and the linear pH range of the electrodes, the electrode prepared with the membrane containing 3% ionophore, 67% o-NPOE and 30% PVC demonstrated the best Nernstian slope of 58.7 71.1 mV pH 1 in the widest pH range
672
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
1.9–12.7 (electrode 4). Hence, further studies were performed with the electrode 4.
500 (a)
400
y = -58.888x + 566.6 R² = 0.9999
3.2. The effect of conditioning and inner filling solution
(b)
300
y = -59.842x + 501.53 R² = 0.994
E, mV
200 100 0 -100 -200 -300
2
4
6
8
10
12
14
pH Fig. 3. Typical calibration graphs of the proposed pH sensing membrane electrode (electrode 4) plotted by the data obtained from (a) pH-ion meter and (b) potentialtime graph.
400 1.96
2.99
300
3.97
4.89 5.92
200
E, mV
The effect of the conditioning upon the performance of the electrode was investigated by the use of two electrodes with the same membrane composition and 1.0 10 3 M CaCl2 as inner filling solution. The electrodes were kept in 1.0 10 3 M HCl and in deionized water for 12 h, separately. The results obtained revealed that the performance and the lifetime of the electrode conditioned in deionized water were higher than that of the electrode conditioned in acid solution. Therefore, the electrodes were kept in deionized water prior to each experiment. The composition and the concentration of the inner filling solution were found to have a significant effect on the potentiometric response of the PVC membrane electrodes by means of changing its selectivity and working range [45]. Either the chloride salts of inert electrolyte or the chloride salt of analyte ion with lower concentration were commonly used as internal solutions [46,47]. For this purpose, the electrode prepared with the membrane containing 3% ionophore, 67% o-NPOE and 30% PVC was filled with 1.0 10 3 M HCl and 1.0 10 3 M CaCl2. From the view point of linear working range, it was determined that the most appropriate inner filling solution was 1.0 10 3 M CaCl2. This application also confirmed that inner filling solutions prepared with inert salts have great effect on the working range of the electrodes [47].
100
6.61 7.84 8.66
0
3.3. Working range and the slope of the electrode
-100
In order to investigate the working range and the slope of the electrode under study, two different types of calibration graphs were plotted (Fig. 3). The potentials measured were plotted against the pH values to obtain the first calibration curve (Fig. 3a) by using series of buffer solutions in the range of pH 1.9–12.9. The second calibration curve (Fig. 3b) was performed by recording the time elapsed to reach a stable potential value after the addition of diluted NaOH to access the appropriate pH (see Fig. 4). The proposed electrode was found much more suitable than many of the other alternatives given in the literature as regard to the linear working range and slope [1,6–14,16,18–20,22– 25,30,32,33]. The comparison of response characteristics of some different type of pH sensing electrodes together with the proposed electrode are given in Table 2. In this study, the impedance measurements were evaluated in two different buffer solutions containing sodium chloride. Fig. 5 (A) displays the impedance plots of the proposed pH sensing membrane and the same membrane without ionophore immersed into acetic acid/acetate buffer-NaCl solution (pH: 4). This type of spectra shows the behaviour of the membranes in the electrodes. As it can be seen, at high frequencies, the signal is mainly dominated by the electrical resistance and the geometric capacitance in parallel with the membrane which showed well defined semi-cyclic impedances [1,7]. The electrical resistance for the ionophore free membrane (electrode 1) was much lower than the resistance for the ionophore-containing membrane (electrode 4). On the contrary, it was stated in literature that the addition of neutral carriers such as alkyldibenzylamines had no effect on the electrical resistance of the membrane electrodes [24]. This behaviour may be attributed to the introduction of a huge neutral macrocyclic compound as calix[4]arene derivative to the membrane. On the other hand, there is a diffusional component ( 451 line) at low frequencies which clearly indicates the reversible diffusion of the hydrogen ion from the solution into the proposed electrode membrane [7].
-200
9.80 11.30 12.22
-300 0
100
200
300
400
500
600
700
800
Time, sec Fig. 4. Dynamic response of the proposed electrode for step changes in pH (electrode 4).
The impedance behavior of the proposed electrode in different pH buffered solutions (pH 3, pH 5 and pH 7) was presented in Fig. 5(B). The electrode displayed nearly the same semi-cyclic impedances at high frequencies. However, the bulk resistance increased with increasing pH values. This phenomenon was related to the transfer of hydrogen ion from solution to the calix [4]arene derivative in polymer layer. Hence, it is in accordance with the literature proposing that the energy of transfer of the hydrogen ion across the polymer membrane increased with increasing pH values [12,13]. 3.4. Response time and lifetime of the electrode The response time of an ion-selective electrode is an important factor for any analytical application. Therefore, the response time required for the electrode to reach 95% of equilibrium mV values after successful immersion in a series of solution each having a 10fold difference in hydrogen ion concentration was measured. The static response time thus obtained was 6–7 s over the entire pH range (1.9–12.7) and no change was observed in 10 min. This was also confirmed from the potential-time graph in Fig. 4. The response times are comparable to, or even shorter than those reported in literature, that were found to give linear response in the same or narrower working ranges [6–12,30–34]. The lifetime of the electrode was determined by reading its potentials and plotting the calibration curves for a period of 12
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
673
Table 2 Comparison of the proposed electrode with some pH sensing electrodes. Electrode
Sensing material
Linear range (pH)
Sensitivity (mV pH 1)
Response time (t95)
References
RSE SCE CWE FET MOE PME PME PME
Quinhydrone Decamethylcyclopentasiloxane N,N-dioctadecylmethylamine Carbon-nanotube thin film MoO3 nanorod Calix[4]-aza-crown derivative Tri-n-dodecylamine Calix[4]arene derivative
2.0–9.5 1.9–9.8 4.0–11.0 3.0–13.0 1.0–5.0 2.5–11.5 4.0–10.0 1.9–12.7
57.5 70.2 57.6 70.2 50.2 7 0.4 50.9 547 2 59.2 547 2 58.7 7 1.1
o 10 s o 15 s o5 s
[1] [6] [14] [16] [18] [30] [22] This work
o 60 s o 10 s o6 s 6–7 s
RSE redox system based electrode; SCE solid contact electrode; CWE coated wire electrode; FET field effect transistor; MOE metal oxide electrode; PME polymeric membrane electrode.
0.8
a b
1.0 0.8
-Z'', MΩ
-Z′′, MΩ
0.6
0.4
0.2
0.6 0.4 0.2 0.0
0.0 0.0
c d e
0.5
1.0
1.5
2.0
2.5
Z′, MΩ
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Z', MΩ
Fig. 5. Typical impedance plots of the membranes containing o-NPOE and PVC (A) in acetic acid/acetate buffer–NaCl solution (pH: 4): (a) without ionophore, electrode 1, (b) with 3% ionophore, electrode 4; (B) in BR buffer-NaCl solutions at: (c) pH: 3, (d) pH: 5, (e) pH: 7 for electrode 4.
months. The slope of the electrode was observed to show no significant change even after this period. This shows that the lifetime of the electrode prepared by the use of 5,17-bis(4benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4] arene was longer than the PVC membrane and other non-glass pH electrodes reported in literature with good repeatability and reproducibility [12,20,22,30–32]. 3.5. The selectivity of the electrode It is reported in literature that the working range of macrocyclic compounds-based electrodes was adversely affected by the anions and the cations. It is known that cations decrease the lower detection limit (at high pH) and anions decrease the upper detection limit (at low pH) of the pH electrodes [23,33]. In short, many ionic impurities are expected to narrow the working range of the electrode [13,43]. The interferences of cations such as Na þ , K þ , Li þ , NH4þ , Mg2 þ , Ca2 þ (0.1 M) and anions such as bromide, iodide, chloride, nitrate, thiocyanate, perchlorate (1.0 10 3 M) were evaluated by the use of the fixed interference method (FIM). The effect of the above mentioned anions on the pH response of the electrode was demonstrated in Fig. 6a. Tentatively notwithstanding, in the presence of higher anion concentrations at low pH values, anions form ion pair with hydrogen ions which are bonded to calix [4]arene thereby making the ionophore irresponsible, for the protons, to penetrate into and the system behaves insensitive to hydrogen ions. In this case, the potential is independent of the pH and is a function of the concentration of anions in the solution [31,32]. As it can be seen in Fig. 6a, the anions except from universal impurity chloride affected the working range of the proposed pH sensing membrane electrode. The following sequence of anion interference was observed: ClO4 4SCN 4I 4NO3 4Br .
Cation interference was also investigated in the region of high pH, where the hydrogen ion concentration is very low, but the above mentioned cation concentration is high, that causes alkaline error in case of some glass pH electrodes. The potentiometric response of the electrode in the presence of these cations indicated that the working range and the Nernstian slope were not changed (Fig. 6b). These results indicated that the corresponding cations would not cause any significant interference unless they present in concentrations 40.1 M. These results are comparable to and even better than some of the data given for other PVC-based membrane electrodes in the literature [6,14,31–33]. Therefore, this electrode can conveniently be used for pH measurements in the media containing these ions.
3.6. The analytical applications of the electrode The electrode was used in the titrations of hydrochloric, hydrofluoric and phosphoric acids with standardized sodium hydroxide. The titration curves were plotted in order to elucidate whether the prepared pH sensing membrane electrode can be employed as an indicator electrode in acid–base titrations. The same titrations were also carried out by the use of a commercial combined glass pH electrode. The expected end-points were observed identical for the proposed electrode and combined glass pH electrode. An example of each titration curves is given as supplementary material (Fig. 7). Furthermore, no erosion effect of hydrofluoric acid was observed on the prepared membrane. After a measurement with this electrode in hydrofluoric acid, the working range and the Nernstian slope remained the same. It can be concluded that the electrode prepared could be successfully employed in acid–base titrations and in fluoride containing media.
674
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
500
500 + H Br I Cl
300
E, mV
300
NO3 ClO4 SCN
200 100 0
+ H + Na + K
400
+ NH4 + Li 2+ Mg 2+ Ca
200
E, mV
400
100 0 -100
-100
-200
-200
-300 0
2
4
6
8
10
12
14
0
2
4
pH
6
8
10
12
14
pH
Fig. 6. Potentiometric response of the proposed electrode in the presence of (a) anions (C ¼1.0 10 3 M), (b) cations (C¼ 0.1 M).
Table 3 The titratable total acidity of beverages found with the proposed electrode and commercial combined glass pH electrode (95% CL, N ¼5, tcrit ¼ 2.78, Fcrit ¼ 6.39).a Beverages
Glass pH electrode
Proposed electrode
texp
Fexp
Orange juice Apple juice Fizzy drink Beer Red wine Vinegar
13.3 7 0.6 9.8 7 0.3 0.6 7 0.0 0.6 7 0.0 8.4 7 0.2 85.3 7 4.3
13.3 7 0.6 9.8 7 0.1 0.7 7 0.0 0.6 7 0.0 8.17 0.2 83.6 7 1.9
0.1 0.4 2.3 0.1 2.1 0.9
1.0 0.2 0.2 1.0 0.4 0.2
a
The results were rounded as one digit beyond the decimal point.
The feasibility of the proposed electrode for the determination of the titratable acidity of some commercial beverages was also investigated [48]. Packaged of beverages were obtained from a local market (Ankara, Turkey). The analytical results obtained by this pH electrode were compared with those obtained by commercial combined glass pH electrode (Table 3). From Table 3, the critical value of t is 2.78 for the 95% confidence level and 4 degree of freedom. Since t4texp, no significant differences between the results obtained with the proposed pH sensing PVC membrane electrode and the commercial combined glass pH electrode was detected. Furthermore, since critical value of F(6.39) 4 Fexp, we concluded that the results obtained from the proposed electrode appear to give better precision than those from combined glass pH electrode at the 95% confidence level. It can be suggested that the developed calix[4]arene based PVC membrane electrode could be an effective tool for potentiometric determination of the titratable acidity in real samples.
4. Conclusions 5,17-Bis(4-benzylpiperidine-1-yl)methyl-25,26,27,28-tetrahydroxy calix[4]arene can be successfully used as an ionophore in pHsensing membrane electrode. The proposed electrode can be a good alternative for a glass electrode. It showed good selectivity for H þ ions in the presence of some cations and anions and a longer lifetime of at least 12 months. This electrode was also found much more suitable when compared with the other PVC membrane pH electrodes reported in the literature as regard to the linear working range and slope. Due to the fact that the electrode had a wide working pH range, it can be not only employed as an indicator electrode in acid–base titrations, but also can be used for the titratable acidity in different real samples. It was also
successfully utilized in fluoride-containing media. The advantages of this electrode are its simplicity of preparation, low-cost, fast response time and long lifetime. Furhermore, it can be possible to prepare a PVC micro-pH electrode using the new calix[4]arene derivative to perform in-vivo pH measurements. We predicted that the proposed electrode could also be used as an anionselective electrode for some anions due to the ability of protonated calix[4]arene binding anions reversibly to the polymer membrane by hydrogen bonds [49].
Acknowledgements We gratefully acknowledge the financial support of Ankara University Research Fund (Grant no: 10B4240003).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.10.032. References [1] J. Ping, Y. Wang, J. Wu, Y. Ying, F. Ji, Microchim. Acta 176 (2012) 229–234. [2] I.M. Illyas, H.H. Mohd, S.S. Intan, A.G Sulaiman, Int. J. Electrochem. Sci. 7 (2012) 12045–12053. [3] X. Ge, Y. Kostov, L. Tolosa, G. Rao, Anal. Chim. Acta 734 (2012) 79–87. [4] R. Ansari, M. Arvand, L. Heydari, J. Chem. Sci. 126 (2014) 41–48. [5] L.B. Peng, L.Y. Heng, S.A. Hasbullah, M. Ahmad, J. Anal. Chem. 62 (2007) 884–888. [6] B. Kim, J. Shim, K.C. Chung, Anal. Lett. 44 (2011) 2138–2149. [7] G.A. Crespo, D. Gugsa, S. Macho, F.X. Rius, Anal. Bioanal. Chem. 395 (2009) 2371–2376. [8] A. Michalska, A. Hulanicki, A. Lewenstam, Analyst 119 (1994) 2417–2420. [9] R.C. Faria, L.O.S. Bulhões, Anal. Chim. Acta 377 (1998) 21–27. [10] N. Zine, J. Bausells, A. Ivorra, J. Aguiló, M. Zabala, F. Teixidor, C. Masalles, C. Viñas, A. Errachid, Sens. Actuators, B 91 (2003) 76–82. [11] W.S. Han, K.C. Chung, M.H. Kim, H.B. Ko, Y.H. Lee, T.K. Hong, Anal. Sci. 20 (2004) 1419–1422. [12] W.S. Han, M.Y. Park, K.C. Chung, K.C. Cho, T.K. Hong, Electroanal 13 (2001) 955–959. [13] W.S. Han, M.Y. Park, K.C. Chung, K.C. Cho, T.K. Hong, Talanta 54 (2001) 153–159. [14] P.W. Alexander, T. Dimitrakopoulos, D.B. Hibbert, Electroanalysis 9 (1997) 1331–1336. [15] J.H. Ahn, J.Y. Kim, M.L. Seol, D.J. Baek, Z. Guo, C.H. Kim, S.J. Choi, Y.K. Choi, Appl. Phys. Lett. 102 (2013) 1–5. [16] Y.S. Chien, W.L. Tsai, I.C. Lee, J.C. Chou, H.C. Cheng, IEEE Electron Device Lett. 33 (2012) 1622–1624. [17] M. Kashif, M.E. Ali, S.M.U. Ali, U. Hashim, S.B.A. Hamid, Nanoscale Res. Lett. 8 (2013) 68–77. [18] G.S. Zakharova, N.V. Podval’naya, J. Anal. Chem. 68 (2013) 50–56.
H.E. Kormalı Ertürün et al. / Talanta 132 (2015) 669–675
[19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]
V.M. Lutov, K.N. Mikhelson, Sens. Actuators, B 18 (1994) 400–403. M. Arvand, K. Ghaiuri, Talanta 79 (2009) 863–870. J. Chojnacki, J.F. Biernat, J. Electroanal. Chem. 277 (1990) 159–164. K. Nurminen, L. Outinen-Y, S. Narkilahti, J. Lekkala, Procedia Eng. 5 (2010) 544–547. D.H. Cho, K.C. Chung, S.S. Jeong, M.Y. Park, Talanta 51 (2000) 761–767. D.H. Cho, K.C. Chung, M.Y. Park, Talanta 47 (1998) 815–821. K.I. Joung, H.J. Yoon, H. Nam, K.J. Paeng, Microchem. J. 68 (2001) 115–120. G.A. Crespo, M.G. Afshar, E. Bakker, Anal. Chem. 84 (2012) 10165–10169. E. Lindner, V.V. Cosofret, R.P. Kusy, R.P. Buck, Talanta 40 (1993) 957–967. J. Langmaier, E. Lindner, Anal. Chim. Acta 543 (2005) 156–176. R.R. Kothur, J. Hall, B.A. Patel, C.L. Leong, M.G. Boutelle, P.J. Cragg, Chem. Commun. 50 (2014) 852–854. X.J. Liu, B. Peng, F. Liu, Y. Qin, Sens. Actuators, B 125 (2007) 656–663. D. Kuruoğlu, E. Canel, S. Memon, M. Yılmaz, E. Kılıc, Anal. Sci. 19 (2003) 217–221. A. Demirel, A. Doğan, E. Canel, S. Memon, M. Yılmaz, E. Kılıc, Talanta 62 (2004) 123–129. V.V. Egorov, Y.V. Sin’kevich, Talanta 48 (1999) 23–28. C. Espadas-Torre, M.E. Meyerhoff, Anal. Chem. 67 (1995) 3108–3114. E.M. Collins, M.A. McKervey, E. Madigan, M.B. Moran, M. Owens, G. Ferguson, S.J. Harris, J. Chem. Soc., Perkin Trans. 1 (1991) 3137–3142.
675
[36] C.D. Gutsche, K.C. Nam, J. Am. Chem. Soc. 110 (1988) 6153–6162. [37] S. Sayin, F. Ozcan, M. Yilmaz, A. Tor, S. Memon, Y. Cengeloglu, Clean—Soil, Air, Water 38 (2010) 639–648. [38] Y.A. Zolotov, Macrocyclic Compounds in Analytical Chemistry, John Wiley and Sons Ltd., USA, 1997. [39] U. Schaller, E. Bakker, E. Spichiger, E. Pretsch, Anal. Chem. 66 (1994) 391–398. [40] Y.W. Choi, N. Minoura, S.H. Moon, Talanta 66 (2005) 1254–1263. [41] Y.W. Choi, S.H. Moon, Environ. Monit. Assess. 70 (2001) 167–180. [42] M.M.G. Antonisse, D.N. Reinhoudt, Electroanalysis 11 (1999) 1035–1048. [43] S. Amemiya, P. Bühlmann, Y. Umezawa, Anal. Chem. 70 (1998) 445–454. [44] R. Eugster, P.M. Gehrig, W.E. Morf, U.E. Spichiger, W. Simon, Anal. Chem. 63 (1991) 2285–2289. [45] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81–84. [46] J Bobacka, V. Väänänen, A. Lewenstam, A. Ivaska, Talanta 63 (2004) 135–138. [47] E. Bakker, P. Buhlmann, E. Pretsch, Electroanalysis 11 (1999) 915–933. [48] Anonim, TS 1125 ISO 750, Meyve ve Sebze Ürünleri Titre Edilebilir Asitlik Tayini, Turkish Standards Institution, Ankara, 2002 (Translated from ISO 750:1998(EN) Fruit and Vegetable Products—Determination of Titratable Acidity). [49] S. Erden, A. Demirel, S. Memon, M. Yılmaz, E. Canel, E. Kılıç, Sens. Actuators, B 113 (1) (2006) 290–296.