highly selective nd(iii) sensors: novel macrocyclic compounds for potentiometric determination of...

Full Paper

Highly Selective Nd(III) Sensors: Novel Macrocyclic Compoundsfor Potentiometric Determination of Neodymium

Ashok Kumar Singh,* Jitendra Singh, A. K. Jain

Department of Chemistry, Indian Institute“of Technology-Roorkee, Roorkee-247 667, India*e-mail: [email protected]

Received: January 29, 2010Accepted: March 9, 2010

AbstractConductometric studies on the complexation properties of two newly synthesized lariat ethers viz 1,5-di(cyano-ethane)-2,4 :7,8 :13,14-tribenzo-1,5-diaza-9,12-dioxacyclopentadeca-2,7,13-triene (L1) and 1,5-di(cyanoethane)-2,3,4-pyridine-7,8 :13,14-dibenzo-1,3,5-triaza-9,12-dioxa cyclopentadeca-2,7,13-triene (L2) towards various metal ions inacetonitrile solutions revealed the formation of 1 : 1 ligand metal complexation. These compounds were explored asneutral ionophores for the fabrication of Nd3þ selective and sensitive membrane coated graphite electrodes (CGEs).Among all the electrodes prepared, CGEs with membrane composition L1(5%) :NaTPB(3%) :NPOE (57%) :PVC(35%) and L2(5%) : NaTPB(3%): NPOE (53%): PVC (39%) showed best performance. Both the electrodes showedNernstian response towards Nd3þ ions over a wide concentration range with detection limits 3.8� 10�8 mol L�1 and1.6� 10�8 mol L�1 respectively. These electrodes showed a fast response time of <15 s and could be used over a periodof three months without significant divergence in their characteristics. The proposed electrodes revealed very goodselectivity for Nd3þ ions over several ions. However, higher concentration of Co2þ, La3þ, Pr3þ and Yb3þ caused someinterference. The potentiometric response of these electrodes was excellent in the range of pH 3.5 to 7.6 and theycould tolerate up to 20% (v/v) nonaqueous media in the test solutions. These electrodes were used successfully asindicator electrode in the potentiometric titration of Nd3þ against EDTA and also in the quantitative determination ofNd3þ ions from binary mixtures and water samples.

Keywords: Coated graphite electrode, Nd3þ selective sensors, Macrocyclic ligands

DOI: 10.1002/elan.201000080

1. Introduction

Metals are used both for domestic and industrial purposes.As a result they occur in environment and cause harmfuleffects on human and animal health. Now a days apart fromtransition elements, rare earth elements (REEs) are alsoindustrially important and have posed significant challengeto the environmentalists as they are enough toxic. Similar toother rare earths, neodymium is moderately toxic andirritating to eyes and long term exposure to neodymium maycause lung embolisms and cell membrane damage which hasnegative influence on reproduction and function of thenervous system [1, 2]. Neodymium is widely used in thepreparation of colour televisions, fluorescent and energysaving lamps, modern vehicles components, data storingdevices, loudspeakers and in petrol-producing industries.The widespread use of neodymium results its occurrence inmany environmental samples where its estimation is desir-able.

A number of instrumental techniques such as spectro-photometry [3, 4] inductively coupled plasma mass spec-trometry (ICP-MS) [5, 6], ICP-AES [7], isotope dilutionmass spectrometry [8 – 10], neutron activation analysis [11,12] and X-ray fluorescence spectrometry [13] etc. arecommonly used for low-level determination of rare-earth

ions in solution. These procedures are often cumbersome,require expertise and large scale infrastructure back up andtherefore not very appropriate for routine analysis ofenvironmental samples. Therefore, a more convenient,inexpensive and fast method of analysis with minimumchemical manipulation is required. Such characteristics areavailable with ion selective electrodes (ISEs) where pro-vided systems of higher sensitivity and selectivity areavailable. Therefore efforts have been made from time totime to develop good ion sensors.

The membranes of neutral ionophores have been success-fully used as ISEs for many metals [14 – 26]. However theirpotential in the determination of rare earths has not beenfully realized due to nonavailability of suitable ionophores.In recent times macrocyclic chemistry has undergonesignificant expansion which has resulted in designingmacrocyclic ligands with interesting and selective coordi-nation properties, thus opening up the possibilities ofpreparing good ionophores for rare earths. The literatureshows that macrocyclic compounds containing varyingcombination of nitrogen (N), oxygen (O) and sulfur (S) asdonor sites tend to form stable complexes with lanthanides[27]. We have now synthesized two new pendent macro-cyclic ligands (L1 and L2) and preliminary studies show thatthey form complexes with many metals. Further stability

Full Paper

Electroanalysis 2010, 22, No. 20, 2443 – 2452 � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 2443

studies have shown that the two pendents studied in thiswork form strong complexes with neodymium, indicatingthat they possess high affinity for Nd3þ ions as compared toothers. Thus they are potential ionophores for preparingNd3þ selective sensors. The literature shows that there areonly limited reports employing macrocyclic compounds [28,29] and Schiff bases [30 – 33] as ionophores in the prepara-tion of neodymium ISEs. Even these electrodes show shortlinear working concentration range and poor detectionlimit. Therefore a better ion selective electrode needs to bedeveloped for neodymium estimation and the presentcommunication is an attempt in this direction and reportsPVC coated graphite electrodes as Nd3þ selective sensorsbased on two novel ionophores.

2. Experimental

2.1. Reagents

Reagent grade sodium tetraphenylborate (NaTPB), potas-sium tetrakis p-(chloro phenyl)borate (KTpClPB), dibu-tylphthalate (DBP), benzyl acetate (BA), dioctylphthalate(DOP), o-nitrophenyloctylether (o-NPOE), dioctylseba-cate (DBS), acetophenone (AP) tetrahydrofuran (THF),sodium borohydride (NaBH4), sodium hydroxide, potassi-um carbonate and high molecular weight poly(vinyl chlo-ride) were purchased from E. Merck (Germany) and used asreceived. 2,6-diaminopyridine, m-phenylenediamine andacrylonitrile were procured from Across organics (USA).The metal chlorides were of analytical grade and usedwithout further purification and their solutions were pre-pared in doubly distilled water. The solutions were stand-ardized wherever necessary.

2.2. Synthesis

Synthesis scheme of the two macrocyclic ligands is shown inFigure 1. Dialdehyde [1,2-bis(2-carboxyaldehydephenoxy)-ethane] used as the precursor, was synthesized according tothe reported method [34]. Novel lariat ethers were synthe-sized by following the previous procedure [35] and charac-terized as follows.

2.2.1 Synthesis of 1,5-Di(cyanoethane)-2,4 :7,8 :13,14-tri-benzo-1,5-diaza-9,12-dioxacyclopentadeca-2,7,13-tri-ene (L1)

Yield: 49%, mp 147 8C. Anal. calcd. for [C28H28N4O2] (%): C,74.31; H, 6.24; N, 12.38; O, 7.07; Observed (%): C, 73.86; H,6.82; N, 12.45; O, 6.86. 1H NMR (DMSO, 500 MHz) dppm:7.0 – 7.7 (m, H�Ar, 12H), 4.6 (m,�Ar�CH2�N�, 4H), 2.5 (t,�C�CH2�CN, 4H), 3.6 (t, �N�CH2�C�, 4H), 4.5 (t,O�CH2�, 4H). FT-IR (KBr, cm�1): 3048 (aromatic C�Hstr.), 2920 (asymmetric �CH2�str.), 2872 (symmetric�CH2�str.), 2253 (�CN str.).

2.2.2. Synthesis of 1,5-Di(cyanoethane)-2,3,4-pyridine-7,8 : 13,14-dibenzo-1,3,5-triaza-9,12-dioxacyclopenta-deca-2,7,13-triene (L2)

Yield: 52%, mp 169 8C. Anal. calcd. For [C27H27N5O2] (%):C, 71.50; H, 6.00; N, 15.44; O, 7.06 Observed (%): C, 70.86; H,6.08; N, 15.23;O, 7.87. 1H NMR (DMSO, 500 MHz) dppm:6.7 – 7.2 (m, H�Ar, 11H), 4.5 (s,�Ar�CH2�N�, 4H), 2.7 (t,�C�CH2�CN, 4H), 3.5 (t,�N�CH2�C�, 4H), 4.3 (t, O�H2�,4H). FT-IR (KBr, cm�1): 3042 (aromatic C�H str.), 2924(asymmetric�CH2�str.), 2863 (symmetric�CH2�str.), 2248(�CN str.).

2.3. Electrode Preparation

The membrane ingredients (ionophore, anion additives,PVC and plasticizers) were dissolved in 5 mL of tetrahy-drofuran and the solvent was partly evaporated to obtain aconcentrated mixture. The viscous solution was thenimmobilized on the spectroscopic grade graphite rods of10 mm length and 5 mm diameter with a fixed copper wire atthe top, by dipping the rod into the solution up to 5 mmdepth, withdrawn quickly and holding it upside-down for afew seconds to allow the THF to dry out. The electrode wassealed into a PVC tube of about the same diameter withepoxy resin. The process was repeated several times until auniform coating formed on the graphite surface and theelectrode was allowed to stabilize overnight. The rod wascovered with paraffin film, keeping the exposed area of themembrane (5 mm) and the contact point open.

2.4. Equilibration of Membranes and PotentialMeasurement

The time of contact and concentration of equilibratingsolution were optimized in order to get stable and reprodu-cible potentials at relatively short response time. Eachelectrode was pre-conditioned before potentiometric meas-urements by soaking it in 1.0� 10�3 M NdCl3 solution for24 h prior to use. The potentials were measured by varyingthe concentration of NdCl3 solution in the range 1.0� 10�1

to 1.0� 10�9 mol L�1. Standard NdCl3 solutions wereobtained by gradual dilution of 0.1 mol L�1 NdCl3 solution.EMF measurements with the coated graphite electrodewere carried out on Century CP 901 digital pH meter at 25�0.1 8C using saturated calomel electrode (SCE) as referenceelectrode by setting up the following cell:

Coated graphite electrode j jSample solution j jHg/HgCl2

jKCl (satd.)

Activity coefficients were calculated according to theDebye – H�ckel procedure, using following equation

log g¼�0.511 z2 [m1/2/(1þ 1.5 m1/2)� 0.2 m] (1)

where m is the ionic strength and z the valency.

Full Paper A. K. Singh et al.

2444 www.electroanalysis.wiley-vch.de � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Electroanalysis 2010, 22, No. 20, 2443 – 2452

www.electroanalysis.wiley-vch.de

2.5. Methodology

2.5.1. Conductometric Study

The complexation of L1 and L2 with a number of cations wasinvestigated conductometrically in acetonitrile solution, at25� 0.05 8C. For this purpose, 25 mL of 1.0� 10�4 mol L�1

cation solution was titrated against a 1.0� 10�2 mol L�1

ionophore solution in acetonitrile media. The conductanceof the solution was measured after each addition of theionophore until the titration plot showed a break.

2.5.2. Determination of Stability Constants

Formation constant of the metal – ionophore complex with-in the membrane phase is an important parameter thatindicates the practical selectivity of the sensor. The ion-ionophore complex formation constants were evaluated bya potentiometric method [36]. In this method, a sandwichmembrane is prepared by fusing two membranes, with only

one containing the ionophore. This membrane electrodewas brought in contact with the aqueous ion solution, havingidentical concentration on both sides, and the cell potentialwas measured. On the other hand the cell potential ofanother membrane having no ionophore was measured. Asreported in the method, the membrane potential (EM) isdetermined by subtracting the cell potential for a membranewithout ionophore from that for the sandwich membrane.The formation constant is then calculated from the follow-ing equation:

bILn¼ LT �

nRT

ZI

� ��n

expEMzIF

RT

� �ð2Þ

where LT is the total concentration of ionophore in themembrane segment, RT is the concentration of lipophilicionic site additives, n is the ion – ionophore complexstoichiometry, and R, T and F are constant having theirusual meaning. zI is the charge on the ion I.

Fig. 1. Reaction scheme for the synthesis of L1 and L2.

Determination of Neodymium

Electroanalysis 2010, 22, No. 20, 2443 – 2452 � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.electroanalysis.wiley-vch.de 2445


2.5.3. pH Measurement

The pH dependence of the electrodes was determined bymeasuring the potential response of 1.0� 10�3 mol L�1 Nd3þ

ion solution as a function of pH in the range 1.0 to 10.0. ThepH was adjusted by using 0.1 M solutions of HCl and NaOH.Thermo Orion 3 star pH benchtop was used to measure thepH of the solutions.

2.5.4. Measurement of Response Time and Life Time

The response time of the CGEs was measured by dipping theelectrode in the test solutions, having a 10-fold increase inconcentration, successively ranging from 1.0� 10�2 to 1.0�10�6 mol L�1 and each time the cell potential was measured.A similar procedure was adopted, in a sequence of high tolow sample concentration, in order to evaluate the reversi-bility of the sensor. In order to determine the durability ofthe electrodes, they were tested daily over a period of4 months. However when not in use, the electrodes werestored in 0.001 M NdCl3 solution.

2.5.5. Selectivity Determination

Selectivity coefficients were determined by IUPAC recom-mended fixed interference method (FIM). Following thefixed interference method, a fixed concentration of inter-fering ion (aB¼ 1.0� 10�2 mol L�1) was added to theprimary Nd3þ ion solutions ranging from 1.0� 10�8 to1.0� 10�2 mol L�1 and the potentials were measured. Thepotential values obtained were plotted versus the activity ofthe primary ion. Potentiometric selectivity coefficients weredetermined graphically using the expression [37]:

KPotNd;B ¼

aNd3þ

ðaBÞzNd

zNdzB

ð3Þ

2.5.6. Aqueous Layer Test

The formation of aqueous layer between the membrane andthe solid contact was tested as introduced by E. Lindner [38].The best performing electrodes were modified by applyingself assembled monolayer of a redox active compound (4-mercaptobenzoic acid) on the graphite surface before castingthe membrane. These electrodes were sequentially exposedto concentrated solutions (c¼0.1 mol L�1) of Nd3þ and La3þ

ions and the potentials were recorded as a function of time.

3. Results and Discussion

The variation of conductance as a function of ligand to metalratio is shown in Figure 2. It is seen from the figure that theconductance variation for Tb, Sm, Gd, Dy is minimalindicating that either these metals cannot be form complexwith the ligands or the formed complex is of weak stability.On the other hand the conductance titration plots for Nd, Pr,

La, show a significant variation in conductance and giving asharp break which gives the molar ratio for the interaction ofmetal ions with these ligands. From these plots it is clear thatthe metal-ligand stoichiometry for these complexes is 1 :1.

The formation constants of the resulting 1 : 1 metal-ionophore complexes were then calculated according tosandwich membrane method. The results, gathered inTable 1, show that the two ligands form most stable complexwith Nd3þ ion having formation constants 8.72 and 9.08 forL1 and L2 respectively. The formation constant values forother metals are much smaller. The higher stability constant

Fig. 2. Conductometric study of ion-ionophore complexes (a)with L1 (b) with L2.




of Nd3þ complexes with L1 and L2 is an indication that bothligands have high affinity for Nd3þ ions compared to othermetals. Therefore these ligands are potential ionophores forpreparing Nd3þ ion selective electrodes. Thus subsequentresults show that the coated graphite electrodes of L1 and L2

act as efficiently selective Nd3þ sensors.

3.1. Optimization of Membrane Composition

It is well known that the membrane composition of ionselective electrodes significantly affects their performance.Therefore a number of electrodes having different compo-sition were prepared and studied. The composition of themembranes was varied by taking different ingredients andchanging their relative amounts. The potential response ofthe coated graphite electrodes containing only ionophoresand PVC was first investigated and the results are summar-ized in Tables 2 and 3. In these studies the concentration ofionophores was varied in the range 2 – 8% (w/w) and foundthat the electrodes having ionophore concentration 5% (w/w) gave the best performance. It is seen from Tables 2 and 3that the electrode nos. 1 and 16 respond to Nd3þ ions linearlyin the concentration range 7.1� 10�6 – 5.0� 10�2 and 3.1�10�6 – 1.0� 10�2 mol L�1 with sub-Nernstian slope 15.9 and16.6 mV decade�1, respectively. Next it was thought neces-sary to improve the performance of the electrodes withregard to higher slope and wider working concentrationrange. It is established [39] that the addition of plasticizergenerally improves the performance of ion selective elec-trode systems due to their influence on the dielectricconstant of the membrane phase. Therefore, in this study,the effect of addition of six plasticizers viz o-NPOE, BA, AP,DBP, DOP and DBS was checked into and the responsesobtained with the plasticized membranes are given inFigure 2 and the results gathered in Tables 2 and 3. It isclear from the performance characteristics of the electrode

nos. 2 to 7 (Table 2) and electrode nos. 17 to 22 (Table 3) thatthe addition of plasticizers generally increases the slope andwidens the working concentration range. The best effect isobtained for o-NPOE as the electrode nos. 2 and 17 havingNPOE plasticized membranes give the wider linear range,1.0� 10�7 – 5.0� 10�2 and 1.6� 10�7 – 1.0� 10�1 mol L�1

with near Nernstian slope 18.0 and 18.8 mV decade�1,respectively. Therefore, in further studies NPOE was used asplasticizer. The better performance of NPOE plasticizedelectrodes (CGE-2 and 17) could be due to high dielectricconstant of the plasticizer which helped to increase thepolarity of the membrane and to reduce the leaching ofmembrane components.

The improvement in the performance of these electrodeswas further attempted by adding lipophilic anion additiveswhich generally suppress the interference by sample anions[40]. The effect of two additives NaTPB and KTpClPB waslooked into the potentiometric response of Nd(III) selectiveelectrodes and the results showed that both the additivesimproved the performance of these CGEs. However of the twoadditives NaTPB performs better and CGE nos. 9 and 24,containing 3% (w/w) NaTPB, gave the best performance of allthe membranes studied so far in terms of widest concentrationrange with Nernstian slope and low detection limit.

Thus at this stage the studies have shown that theingredients producing best membranes along with iono-phores (L1 and L2) are additive NaTPB, Plasticizer o-NPOEand PVC as inert Matrix. Further improvisation in thepotentiometric characteristics of CGE-9 and 24 was tried bychanging the amount of ionophores only. The results showthat the presence of higher amounts of ionophore does notin anyway enhance the performance of these electrodes butrather adversely affect the performance by decreasing theslope and shortening the linear range. Thus the optimumamount of the ionophores in these studies is found to be 5%(w/w). From these studies it is reasonable to conclude thatthe optimum compositions L1 : NPOE :NaTPB :PVC as5 :57 :3 :35 (%, w/w) (CGE-9) and L2:NPOE:NaTPB:PVCratio as 5 : 53 : 3 :39 (%, w/w) (CGE-24) perform best in allrespects. Further it was observed from aqueous layer testthat there was no drift in the potentials with time. However,stable potential values were obtained for SAM modifiedgraphite electrodes. Therefore the possibility of the forma-tion of undesirable aqueous layer between the ion selectivemembrane and its solid contact can be neglected. Thereforeall further studies were carried out with CGE-9 and 24.

3.2. Effect of pH on the Performance of the Electrodes

The effect of pH on the performance of the two coatedgraphite electrodes was investigated by measuring thepotential in a 1.0� 10�3 mol L�1 NdCl3 solution at differentpH values from 1 to 10. The pH was adjusted with 0.1 MHCl/NaOH and the potential response obtained at differentpH are shown in Figure 4. It is seen from this figure that thepotentials remain constant in the pH range 3.5 to 7.6. At pHvalue above 7.6, the potentials start decreasing which appear

Table 1. Formation constants of different metal complexes withionophores L1 and L2. s : Standard deviation of three measure-ments.

Metal ions (Mnþ) Stability constants

(log bML1)� s (log bML2)� s

Nd3þ 8.72� 0.4 9.08� 0.3La3þ 3.80� 0.4 3.26� 0.7Yb3þ 3.61� 0.5 3.49� 0.5Pr3þ 3.37� 0.4 3.32� 0.7Tb3þ 2.12� 1.1 2.67� 0.8Sm3þ 2.48� 0.4 2.32� 0.4Dy3þ 2.31� 1.3 2.19� 1.6Gd3þ 2.06� 0.9 2.16� 0.4Co2þ 5.14� 0.2 5.38� 0.4Cu2þ 2.02� 0.9 2.04� 1.2Ni2þ 1.96� 0.6 2.13� 1.4Ca2þ 1.29� 0.3 1.48� 0.8Naþ 1.05� 0.8 1.66� 0.3Agþ 1.16� 0.6 1.28� 0.4




probably due to the formation of metal hydroxide species inthe system [41]. Similarly in the acidic range, i.e., at pH lessthan 3.5, the response of these electrodes increased ratherirregularly with decreasing the pH of the solution. At suchhigh acidities, the membrane may extract Hþ ions in additionto Nd(III) ions [42] due to the protonation of donor sites ofthe ligands. Thus these Nd(III) selective CGEs were foundto perform successfully in the pH range 3.5 – 7.6.

3.3. Potentiometric Selectivity

The selectivity is one of the most important characteristics ofan ion sensor which determines its utility for analyticalpurpose. The potentiometric selectivity coefficient valuesfor the electrodes (CGE-9 and 24) were determinedaccording to FIM and the results obtained are reported inTable 4. The values of selectivity coefficients show that these

electrodes are selective to Nd3þover monovalent (Agþ, Naþ,Kþ , Liþ) and divalent (Hg2þ, Fe2þ, Co2þ, Ni2þ, Cd2þ, Mg2þ,Ca2þ, Zn2þ, Cu2þ) ions. The stability of the Nd complex of thetwo lariat ethers is accountable for selectivity of theseelectrodes towards Nd3þ over other interfering ions. How-ever the selectivity coefficient values for cobalt are not verysmall and thus, estimation of neodymium can be done onlyin presence of small amount of Co2þ. In order to estimate theextent of interference caused by cobalt, the performance ofthe two electrodes was checked in presence of differentconcentrations of Co2þ and the results showed that the twoelectrodes show no significant change in their performancecharacteristics at lower concentrations of Co2þ ion. Whenthe concentration of Co2þ is increased up to 1.0� 10�2 molL�1, interference caused a shortening of linear range andelevation in detection limits with increased value of slope.However it was found that the presence of 5.0� 10�3 molL�1 Co2þ can be tolerated over the entire concentration

Table 2. Potentiometric response characteristics of the electrodes based on L1.

CGEs Membrane composition (% w/w) Linear range(mol L�1)

Slope(mV decade�1)

Detection limit(mol L�1)

L1 Additive Plasticizer PVC

1 5 – – 95 7.1� 10�6 – 5.0� 10�2 15.9 4.4� 10�6

2 5 – 57, NPOE 38 1.0� 10�7 – 5.0� 10�2 18.0 8.0� 10�8

3 5 – 57, AP 38 9.4� 10�7 – 1.0� 10�1 17.3 5.0� 10�7

4 5 – 57, DBP 38 1.7� 10�6 – 1.0� 10�2 16.9 7.9� 10�7

5 5 – 57, DOP 38 6.2� 10�6 – 3.1� 10�2 16.6 2.8� 10�6

6 5 – 57, BA 38 7.4� 10�6 – 1.0� 10�1 16.5 3.1� 10�6

7 5 – 57, DBS 38 5.8� 10�6 – 1.0� 10�1 16.2 4.2� 10�6

8 5 3, KTpClPB 57, NPOE 35 3.7� 10�7 – 1.0� 10�2 18.4 1.3� 10�7

9 5 3, NaTPB 57, NPOE 35 8.4� 10�8 – 3.1� 10�2 19.8 3.8� 10�8

10 5 3.5, NaTPB 57, NPOE 34.5 8.4� 10�8 – 1.0� 10�2 20.3 6.6� 10�8

11 5 4, NaTPB 57, NPOE 34 5.6� 10�7 – 1.0� 10�2 22.5 1.2� 10�7

12 4 3, NaTPB 57, NPOE 36 2.2� 10�6 – 5.0� 10�2 18.3 6.8� 10�7

13 6 3, NaTPB 57, NPOE 34 1.8� 10�7 – 1.0� 10�2 19.2 9.2� 10�8

14 8 3, NaTPB 57, NPOE 32 7.2� 10�7 – 1.0� 10�1 18.6 3.7� 10�7

15 – 3, NaTPB 57, NPOE 40 3.1� 10�3 – 1.0� 10�1 6.6 8.2� 10�4

Table 3. Potentiometric response characteristics of the electrodes based on L2.

CGEs Membrane composition (% w/w) Linear range(mol L�1)

Slope(mV decade�1 )


L2 Additive Plasticizer PVC

16 5 – – 95 3.1� 10�6 – 1.0� 10�2 16.6 6.3� 10�7

17 5 – 53, NPOE 42 1.6� 10�7 – 1.0� 10�1 18.8 6.8� 10�8

18 5 – 53, AP 42 7.9� 10�7 – 3.1� 10�2 17.8 5.1� 10�7

19 5 – 53, DBP 42 2.4� 10�6 – 3.1� 10�2 17.1 1.0� 10�6

20 5 – 53, BA 42 5.0� 10�6 – 1.0� 10�1 17.3 4.2� 10�6

21 5 – 53, DOP 42 5.2� 10�6 – 5.0� 10�2 17.0 4.0� 10�6

22 5 – 53, DBS 42 7.8� 10�6 – 5.0� 10�2 16.8 4.2� 10�6

23 5 3, KTpClPB 53, NPOE 39 6.2� 10�7 – 1.0� 10�1 18.8 3.9� 10�7

24 5 3, NaTPB 53, NPOE 39 4.6� 10�8 – 5.0� 10�2 19.7 1.6� 10�8

25 5 3.5, NaTPB 53, NPOE 38.5 4.6� 10�8 – 1.0� 10�2 19.9 4.1� 10�8

26 5 4, NaTPB 53, NPOE 38 7.2� 10�7 – 1.0� 10�1 23.5 5.4� 10�7

27 4 3, NaTPB 53, NPOE 40 4.6� 10�7 – 1.0� 10�2 19.1 1.3� 10�7

28 6 3, NaTPB 53, NPOE 38 2.3� 10�7 – 5.0� 10�2 18.9 7.8� 10�8

29 8 3, NaTPB 53, NPOE 36 3.1� 10�7 – 1.0� 10�2 17.4 1.0� 10�7

30 – 3, NaTPB 53, NPOE 44 2.4� 10�3 – 1.0� 10�1 5.9 8.2� 10�4




range for the two electrodes. Further selectivity coefficientvalues indicate that the electrodes are selective to Nd3þ ionover a number of lanthanides (Gd3þ, Pr3þ, Ce3þ, La3þ, Tb3þ,Dy3þ Sm3þ and Yb3þ). However, between the two electrodesCGE-24 based on L2 was found better regarding selectivityas CGE-9 based on L1 which showed significant interferencefor La3þ, Pr3þ and Yb3þ. Thus the two electrodes can be usedfor neodymium estimation in presence of rare earthelements. However, for CGE-9 the concentration of La3þ,Pr3þ and Yb3þ should not exceed than 1.0� 10�2 mol L�1

while estimating neodymium in view of not very highselectivity of this sensor.

3.4. Response Time and Life Time Study

The response time of the electrodes was determined bymeasuring the time required to achieve a steady statepotential (within�1 mV) after successive immersion of theelectrodes in a series of neodymium ion solutions, eachhaving a 10-fold increase in concentration from 1.0� 10�6 to1.0� 10�2 mol L�1. The typical potential – time plots for thetwo electrodes showed that the response time for theelectrode based on L1 is 14 s where as for the electrode basedon L2 it is 10 s. The potential readings remained constant forabout 3 min after which a slow divergence was observed.

The loss of membrane components from the electrode isthe major cause for the short lifetime of the polymermembrane electrodes. The life time of the electrodes wasmeasured by monitoring the change in slope and workingrange with time. The electrodes were used daily over aperiod of 4.5 months and their characteristics were ob-served. The results showed that the two electrodes have along life time and can be used successfully for 3 monthsafter which some change is observed in their performancecharacteristics towards Nd3þ ion due to swelling ofmembrane and leaching of membrane components oncontinuous usage. The reproducibility of the Nd3þ elec-trode was also investigated and the standard deviations of10 replicate measurements at 1.0� 10�3 mol L�1 were�0.3and �0.4 mV, for CGE-9 and 24 based L1 and L2,respectively.

3.5. Effect of Surfactants

Surfactants are highly surface active molecules and affectthe performance of the electrode as they have a tendency to

Fig. 3. Potential responses of Nd3þ ion selective electrodes based on (a) L1 and (b) L2 with different plasticizers.

Fig. 4. Effect of pH on the performance of CGE-9 and CGE-24at 1.0� 10�3 mol L�1 NdCl3 solution.




accumulate and adsorb at the sample membrane interface.In order to evaluate the effect of surfactants on potentio-metric properties of the CGEs based on the ionophores L1

and L2, their basic analytical parameters were determined inthe presence of cationic, TBC (cmc, 1.7� 10�4 M) andanionic, SDS (cmc, 8.0� 10�4 M) surfactants.. The effect ofsurfactants was negligible at concentration below 10�4 M,however above 10�4 M concentration surfactants startadversely affecting the performance of the electrodesshowing decreased slope and working concentration rangeand increased detection limit. Further the selectivity of the

electrodes also decreased in the presence of surfactants. Theinterference is probably due to the adsorption of surfactantat the membrane solution interface.

3.6. Effect of Partially Nonaqueous Mixtures

The performance of the two electrodes was also investigatedin partially nonaqueous media using acetonitrile – water,methanol – water and ethanol – water mixtures and theresults showed that the two electrodes tolerate up to 20%(v/v) nonaqueous content. However at higher nonaqueouscontent the performance of the electrodes started deterio-rating which may be due to the leaching of ionophores fromthe membrane phase. The electrode performance alsoaffected when the electrodes were stored in nonaqueoussolution for one week.

3.7 Comparisons with Reported Sensors

The performance characteristics of the prepared electrodes(CGE-9 and 24) were compared with the other Nd3þ ionselective electrodes based on various ionophores and shownin Table 5. The comparative study of the two CGEs based onL1 and L2 reveals that they are superior over the previouslyreported Nd(III) selective electrodes with regard to theirwide working range and low detection limit. However, theirselectivity is comparable to the reported electrodes.

4. Analytical Applications

The proposed CGEs were applied to the monitoring of Nd3þ

in various binary mixtures, prepared by adding 10 mL of

Table 4. Selectivity coefficients of CGE-9 and 24 with variousinterfering ions.

Interfering ions Selectivity coefficients (� log KNd3þ,B)

CGE-9 CGE-24

La3þ 1.6� 0.1 2.0� 0.2Yb3þ 1.7� 0.1 2.5� 0.1Pr3þ 2.1� 0.1 1.9� 0.1Tb3þ 2.3� 0.2 2.2� 0.4Sm3þ 2.3� 0.1 2.1� 0.2Gd3þ 2.4� 0.2 2.1� 0.1Dy3þ 2.6� 0.3 2.7� 0.1Co2þ 1.1� 0.3 1.2� 0.1Cu2þ 4.1� 0.2 4.3� 0.1Hg2þ 4.3� 0.2 4.4� 0.2Ce3þ 4.3� 0.1 4.5� 0.3Zn2þ 4.4� 0.2 3.3� 0.2Cd2þ 4.3� 0.1 3.2� 0.3Ni2þ 4.2� 0.3 4.1� 0.1Ca2þ 4.5� 0.1 4.0� 0.3Naþ 3.8� 0.3 3.6� 0.2Agþ 3.6� 0.2 3.4� 0.2Kþ 4.1� 0.2 3.6� 0.1Liþ 4.4� 0.1 3.8� 0.1

Table 5. Comparison of Nd3þ selective CGEs with the reported electrode

Ref. No. Slope(mV decade�1)

Linear range(mol L�1)


pH range � log KNd3þ,B (FIM)

CGE-9 19.8� 0.4 8.4� 10�8 – 3.1� 10�2 3.8� 10�8 3.5 – 7.6 La3þ (1.6), Yb3þ (1.7), Tb3þ (2.3), Ce3þ (4.3),Dy3þ (2.6), Gd3þ (2.4), Co2þ (1.1), Zn2þ (4.4),Cu2þ (4.1), Ni2þ (4.2), Agþ (3.6), Naþ (3.8)

CGE-24 19.7� 0.5 4.6� 10�8 – 5.0� 10�2 1.6� 10�8 3.5 – 7.6 La3þ (2.0), Yb3þ (1.9), Tb3þ (2.2), Ce3þ (4.5),Dy3þ (2.7), Gd3þ (2.1), Co2þ (1.2), Zn2þ (3.3),Cu2þ (4.3), Ca2þ (4.0), Kþ (3.6), Naþ (3.6)

28 20.1� 0.2 1.0� 10�6 – 1.0� 10�2 7.9� 10�7 4.0 – 6.5 La3þ (1.5), Yb3þ (1.2), Sm3þ (1.9), Ce3þ (1.3),Gd3þ (1.3), Co2þ (1.7), Zn2þ (1.5), Cu2þ (1.4),Ni2þ (1.9), Agþ (1.4), Naþ (1.4)

30 19.6� 0.3 1.0� 10�5 – 1.0� 10�2 7.0� 10�6 4.0 – 8.0 La3þ (1.7), Yb3þ (2.3), Sm3þ (1.7), Dy3þ (2.3),Gd3þ (1.9), Cu2þ (2.1), Kþ (2.7), Naþ (2.9)

31 19.6� 0.3 1.0� 10�5 – 1.0� 10�2 2.0� 10�6 3.5 – 8.5 La3þ (1.6), Yb3þ (2.2), Sm3þ (1.6), Ce3þ (1.6),Dy3þ (2.1), Gd3þ (1.8), Cu2þ (2.1), Ca2þ (2.4),Kþ (2.6), Naþ (2.7)

32 19.7� 0.4 1.0� 10�6 – 1.0� 10�2 6.2� 10�7 3.7 – 8.3 La3þ (2.1), Tb3þ (2.9), Sm3þ (3.5), Dy3þ (2.0),Gd3þ (2.1), Co2þ (3.3), Zn2þ (3.1), Cu2þ (3.1),Ni2þ (2.9), Agþ (2.2), Naþ (3.0), Hg2þ (2.3)

33 19.8� 0.3 5.0� 10�7 – 1.0� 10�2 1.0� 10�7 4.0 – 8.0 La3þ (3.8), Tb3þ (3.2), Sm3þ (3.8), Dy3þ (4.5),Gd3þ (3.6), Tb3þ (4.4), Co2þ (2.4)




5.0� 10�3 mol L�1 solution of different cations to 1.0�10�4 mol L�1 neodymium ion solution at 5.0 pH. The resultsare summarized in Table 6. The recovery of Nd3þ ions fromthe test samples is satisfactory which shows high selectivityof these electrodes for Nd3þ ions over the interfering ions.

The analytical utility of these electrodes was also eval-uated by carrying out potentiometric titration of 20 mL of5.0� 10�3 mol L�1 Nd3þ ions against 1.0� 10�2 mol L�1

EDTA at pH 5.0. The titration plots (Figure 5) were ofsigmoid shapes that indicates high selectivity of theseelectrodes for Nd3þ.

The proposed electrodes were also used for the monitor-ing of Nd3þ ions in Ganga (Roorkee) and Yamuna (Delhi)river water and also in the waste water taken from thechemistry department of this institute. The potentiometricresults, given in Table 7 are in close agreement with thoseobtained by atomic absorption spectrometry.

5. Conclusions

Nd3þ ion selectivity of two novel pendent macrocycles hasbeen looked into using them as ionophore in the membranecoated graphite electrodes. Among all the electrodesprepared with different membrane compositions, CGE-9and 24 incorporating o-NPOE as solvent mediator andNaTPB as anionic exchanger performed best. Their perfor-mance characteristics towards Nd3þ ion were characterizedas wide working concentration range (8.4� 10�8 – 3.1� 10�2

and 4.6� 10�8 – 5.0� 10�2 mol L�1) with Nernstian slope(19.8� 0.4 and 19.7� 0.5 mV decade�1 of a3þ

NdÞ, low detec-tion limit (3.8� 10�8 and 1.6� 10�8 mol L�1), useful pHrange (3.5 to 7.6), fast response time of 14 s (for CGE-9) and

10 s (for CGE-24) with a shelf life of three months. Theycould also tolerate up to 20% (v/v) nonaqueous content(methanol, ethanol and acetonitrile) in the test solution. Thepresence of surfactants at 1.0� 10�4 mol L�1 or abovecaused significant interference in the performance of the

Table 6. Determination of Nd3þ ions in various binary mixtures by the proposed CGEs. s : Standard deviation of three measurements.

Nd3þ ion concentration (M) Added cation concentration (M) Recovery (%)� s

CGE-9 CGE-24

1.0� 10�4 La3þ (5.0� 10�3) 96.8� 0.3 97.8� 0.41.0� 10�4 Yb3þ (5.0� 10�3) 97.3� 0.3 97.6� 0.31.0� 10�4 Tb3þ (5.0� 10�3) 99.2� 0.5 99.4� 0.61.0� 10�4 Gd3þ (5.0� 10�3) 99.5� 0.1 100.5� 0.81.0� 10�4 Dy3þ (5.0� 10�3) 98.4� 0.9 101.6� 0.21.0� 10�4 Co2þ (5.0� 10�3) 97.4� 0.2 96.9� 0.11.0� 10�4 Cu2þ (5.0� 10�3) 100.2� 0.1 101.5� 0.31.0� 10�4 Hg2þ (5.0� 10�3) 99.6� 0.3 99.1� 0.21.0� 10�4 Zn2þ (5.0� 10�3) 99.3� 0.5 98.8� 0.61.0� 10�4 Cd2þ (5.0� 10�3) 99.5� 0.4 99.2� 0.2

Fig. 5. Potentiometric titration curve for 20 mL of 5.0� 10�3 molL�1 Nd3þ with 1.0� 10�2 mol L�1 EDTA, at constant pH 5.0 usingthe proposed electrodes.

Table 7. Quantitative determination of Nd3þ ions in water samples.

Sample Added Nd3þ ion concentration (ppm) Found by CGEs (ppm) Found by AAS (ppm)

CGE-9 CGE-24

Ganga water 5.00 5.32 5.06 4.97Yamuna water 5.00 5.14 5.10 5.25Waste water 5.00 4.88 5.36 5.20




electrodes. Further the electrodes were found to be effi-ciently selective over a number of metal ions and couldtherefore be used for the analysis of Nd3þ ions in binarysolutions. These electrodes could be used in successfuldetermination of Nd3þ ions from water samples and also asindicator electrodes in the potentiometric titration withEDTA.

Acknowledgement

Jitendra is grateful to The Council of Scientific andIndustrial Research (CSIR), New Delhi, India for providingfinancial assistance for this work.

References

[1] R. E. Kirk, D. F. Othmer, Encyclopedia of Chemical Tech-nology, Wiley, New York 1982.

[2] G. S. S. Rao, V. K. Upadhyay, R. C. Jain, J. Appl. Polym. Sci.1999, 71, 595.

[3] D. B. Gladilovich, V. Kuban, L. Sommer, Talanta 1988, 35,259.

[4] A. Hrdlicka, J. Havel, C. Moreno, M. Valiente, Anal. Sci.1991, 7, 925.

[5] R. S. Houk, V. A. Fassel, G. D. Flesch, H. J. Svec, Anal.Chem. 1980, 52, 2283.

[6] N. Shibata, N. Fudagawa, M. Kubota, Anal. Chem. 1991, 63,636.

[7] A. Mazzucotelli, F. DePaz, E. Magi, R. Frache, Anal. Sci.1992, 8, 189.

[8] A. Masuda, N. Nomura, T. Tanaka, Geochim. Cosmochim.Acta 1973, 37, 239.

[9] M. Dorazio, S. Tonarini, Anal. Chim. Acta 1997, 351, 325.[10] G. L. Ramendik, E. V. Fatyushina, A. I. J. Stepanov, Anal.

Chem. 2003, 58, 152.[11] S. F. Marsh, Anal. Chem. 1967, 39, 641.[12] J. J. Mccown, J. P. Larsen, Anal. Chem. 1961, 33, 1003.[13] D. H. Cornell, Pure Appl. Chem. 1993, 65, 2453.[14] A. F. Zhulkov, D. Erne, D. Ammann, M. Guggi, E. Pretsch,

W. Simon, Anal. Chim. Acta 1981, 131, 117.[15] V. P. Y. Gadzekpo, G. D. Christian, Anal. Lett. 1983, 16, 1371.[16] A. Ohki, J. P. Lu, J. L. Hallman, X. Huang, R. A. Bartsch,

Anal. Chem. 1995, 67, 2405.

[17] E. Bakker, Anal. Chem. 1997, 69, 1061.[18] M. H. Lee, C. L. Yoo, J. S. Lee, C. In-Sook, B. H. Kim, G. S.

Cha, H. Nam, Anal. Chem. 2002, 74, 2603.[19] K. Suzuki, K. Tohda, H. Aruga, M. Matsuzoe, H. Inoue, T.

Shirai, Anal. Chem. 1988, 60, 1714.[20] A. K. Singh, V. K. Gupta, B. Gupta, Anal. Chim. Acta 2007,

585, 171.[21] V. K. Gupta, A. K. Singh, S. Mehtab, B. Gupta, Anal. Chim.

Acta 2006, 566, 5.[22] V. K. Gupta, A. K. Singh, M. A. Khayat, B. Gupta, Anal.

Chim. Acta 2007, 590, 81.[23] R. W. Cattrall, P. Chin-Poh, Anal. Chem. 1976, 48, 552.[24] J. Senkyr, D. Ammann, P. C. Meier, W. E. Morf, E. Pretsch,

W. Simon, Anal. Chem. 1979, 51, 786.[25] D. A. Chowdhury, T. Ogata, S. Kamata, Anal. Chem. 1996,

68, 366.[26] V. K. Gupta, A. K. Singh, B. Gupta, Anal. Chim. Acta 2006,

575, 198.[27] V. Alexandra, Chem. Rev. 1995, 95, 273.[28] M. Shamsipur, M. Hosseini, K. Alizadeh, M. F. Mousavi, A.

Garau, V. Lippolis, A. Yari, Anal. Chem. 2005, 77, 276.[29] S. Chandra, D. V. Singh, Mater. Sci. Eng. A 2009, 502, 107.[30] P. Norouzi, M. R. Ganjali, A. Ahmadalinezhad, M. J. Adib,

Braz. Chem. Soc. 2006, 17, 1309.[31] M. R. Ganjali, A. Ahmadalinezhad, P. Norouzi, M. Adib, J.

Appl. Electrochem. 2006, 36, 931.[32] H. Behmadi, H. A. Zamani, M. R. Ganjali, P. Norouzi,

Electrochim. Acta 2007, 53, 1870.[33] V. K. Gupta, R. N. Goyal, R. A. Sharma, Anal. Chim. Acta

2009, 647, 66.[34] A. Simion, C. Simion, T. Kanda, S. Nagashima, Y. Mitoma, T.

Yamada, K. Mimura, T. J. Masashi, Chem. Soc., Perkin Trans.2001, 1, 2071.

[35] A. K. Singh, P. Singh, S. Banerjee, S. Mehtab, Anal. Chim.Acta 2009, 633, 109.

[36] Y. Mi, E. Bakker, Anal. Chem. 1999, 71, 5279.[37] Y. Umezawa, K. Umezawa, H. Sato, Pure Appl. Chem. 1995,

67, 507.[38] E. Lindner, R. E. Gyurcsanyi, J. Solid State Electrochem.

2009, 13, 51.[39] R. Eugster, T. Rosatzin, B. Rusterholz, B. Aebersold, U.

Pederazza, D. Ruegg, A. Schmid, U. E. Spichiger, W. Simon,Anal. Chim. Acta 1994, 289, 1.

[40] D. Ammann, E. Pretsch, W. Simon, E. Lindner, A. Bezegh,E. Pungor, Anal. Chim. Acta 1985, 171, 119.

[41] W. Wang, Y. Chen, M. Wu, Analyst 1984, 109, 281.[42] S. Kamata, R. Kojima, Anal. Sci. 1994, 10, 409.




highly selective nd(iii) sensors: novel macrocyclic compounds for potentiometric determination of...

Documents