A

2twasrClm©

K

1

tbi[ivtT

n

0d

Talanta 71 (2007) 1986–1992

Cation selectivity of ionophores based on tripodalthiazole derivatives on benzene scaffold

Hong-Seok Kim a,∗, Dong-Hyun Kim a, Ki Soo Kim a, Jun-Hyeak Choi a,Heung-Jin Choi a, Sung-Hoon Kim b, Jun Ho Shim c,

Geun Sig Cha c, Hakhyun Nam c,∗∗a Department of Applied Chemistry, Kyungpook National University, Daegu 702-701, Republic of Korea

b Department of Textile System Engineering, Kyungpook National University, Daegu 702-701, Republic of Koreac Chemical Sensor Research Group, Department of Chemistry, Kwangwoon University, Seoul 139-701, Republic of Korea

Received 31 July 2006; received in revised form 1 September 2006; accepted 1 September 2006Available online 9 October 2006

bstract

The synthesis and potentiometric evaluation of new 1,3,5-tris(thiazolylcarbethoxy)-2,4,6-trimethylbenzene (3), 1,3,5-tris(thiazolylhydroxy)-,4,6-trimethylbenzene (4), 1,3,5-tris(thiazolylmethyl)-2,4,6-trimethylbenzene (5), and 1,3,5-tris(thiazolylphenyl)-2,4,6-trimethylbenzene (6),oward mono and divalent cations under various pH conditions are outlined. The ion-selective properties of the newly synthesized compoundsere studied by measuring the potentiometric responses of the 3-, 4-, 5-, and 6-based membrane electrodes to alkali metal, alkaline earth metal,

mmonium, and transition metal ions, under various pH conditions. The 3-based electrode exhibited a Nernstian response to ammonium and potas-ium under alkaline pH conditions, while the other three electrodes showed a poor potentiometric performance. All electrodes showed substantialesponses to silver ion under acidic condition, but there was almost nil response to other transition metal ions (Fe2+, Co2+, Zn2+, Ni2+, Pb2+, Cd2+,

u2+ and Hg2+). The 3- and 5-based electrodes resulted in near Nernstian responses (51.3 mV and 59.5 mV/pAg+, respectively) with low detection

imits (∼100 ppt), while the 4- and 6-based ones showed sub-Nernstian below 40 mV/pAg+. The results were interpreted with semi-empiricallyodeled structures.2006 Elsevier B.V. All rights reserved.

cm[

rt[a

eywords: Tripodal thiazole scaffold; Ionophore; Cation selectivity; Silver (I)

. Introduction

The design and research of artificial receptors for the selec-ive recognition of metal ions has attracted increasing interestecause of its significant importance and potential applicationn physiological, environmental, and supramolecular chemistry1]. Silver (I) is one of the main transition metals used inndustries and drugs. Due to their antibacterial properties, sil-

er compounds have been used to disinfect potable water, andhey have been used for dental and pharmaceutical purposes.herefore, the design of Ag+-selective neutral carriers and the

∗ Corresponding author. Tel.: +82 53 950 5588; fax: +82 53 950 6594.∗∗ Co-corresponding author. Tel.: +82 2 940 5246; fax: +82 2 942 4635.

E-mail addresses: kimhs@knu.ac.kr (H.-S. Kim),amh@daisy.kw.ac.kr (H. Nam).

tsRtastue

039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved.oi:10.1016/j.talanta.2006.09.001

onstruction of ion-selective electrodes (ISEs) have attracteduch attention regarding environmental and clinical analyses

2,3].A benzene ring may be used as a small, rigid platform for

eceptor systems. Since McNicol discovered the preorganiza-ion of functional groups in hexasubstituted benzene derivatives4], benzene-based receptor molecules have been widely useds building blocks for extended, well-defined molecular archi-ecture and as a scaffold of synthetic receptors that show highelectivity toward cations, anions and organic molecules [5–12].aymond et al. synthesized tripodal ionophores by attaching

hree catechol units to benzene and mesitylene and their cationffinity was evaluated [13]. They found that the binding con-

tant of 1,3,5-tris(catechol)mesitylene for Fe (III) is higher thanhat of the natural compound enderrobactin, which showedniquely strong binding affinity toward Fe (III). This discov-ry has led to the exploration of this benzene motif for cation

H.-S. Kim et al. / Talanta 7

raawtTlNp[

euhcmhr

tmerfapwdm

t[

2

2

caIUrlss(yMpMl(Gud

ewwbaFm2Sl

Fig. 1. Structures of the newly-synthesized ionophores.

ecognition purposes. By incorporating heterocyclic units suchs pyrazol, indole, and pyridine into a benzene motif, complex-tion studies with Pd(II) [14], Fe(II) [15,16], and Cu(I) [17]ere performed. Kim et al. synthesized tris(pyrazol)benzene for

he recognition of NH4+ and alkyl ammonium cations [18,19].

hey found that ion-selective electrodes using the immobi-ized tris(pyrazol)benzene showed profound selectivity towardH4

+ over Na+ and K+. Ahn et al. also studied the com-lexities of alkylammonium cations by tris(oxazolin)benzene11].

Various ionophores including calix[4]arenas [20–22], crownther derivatives [23–25], and steroidal tweezers [26] have beensed for the selective detection of Ag+ ions by introducing softeteroatoms such as N and S, as an electron donor to metalations. In order to enhance their complexing cation selectivity,any heterocyclic units such as pyridyl, bipyridyl, bithiazoloyl

ave been introduced into calixarene on their lower and upperims [20–22].

Recently, we have designed new neutral carriers that containhiazole moieties [27–31]. The ISEs based on solvent polymericembranes (plasticized PVC) doped with these ionophores,

xhibited large cation selectivity. To better understand the cationecognition mechanisms of these membranes, we synthesizedour different mesitylene-based receptors having tripodal thi-zole units (compounds 3–6 in Fig. 1 and Scheme 1), and

repared those receptor-based membrane ISEs. In this paper,e report their response characteristics to various mono andivalent cations (alkali metal, alkaline earth metal, transitionetal ions and ammonium), as well as our understanding of

swb[

Scheme 1. Synthetic procedure of

1 (2007) 1986–1992 1987

heir ion-recognition properties by the use of model structures32].

. Experimental

.1. General

Melting points were determined using a Thomas–Hooverapillary melting point apparatus (Thomas Scientific, USA). 1Hnd 13C NMR spectra were obtained using a Varian UNITYnova 300WB FT-NMR Spectrometer (Varian Inc., Palo Alto,SA) in CDCl3. Chemical shifts in the 1H NMR spectra were

eported using δ units downfield from the internal tetramethylsi-ane. The IR spectra were measured with a Galaxy FT-IR 7000pectrophotometer (Mattson Instrument, Madison, USA). Masspectra were recorded on a Shimadzu QP-1000 spectrometerShimadzu Scientific Instrument, Kyoto, Japan). Elemental anal-ses were performed on a Calro Erba 1106 (CE Instruments,illan, Italy) at the Center for Scientific Instruments, Kyung-

ook National University. TLC analyses were carried out onerck silica gel 60F254 plates, visualized with a 254-nm UV

amp. For routine column chromatography, Merck silica gel70–230 mesh) was used as the adsorbent (Merck, Darmstadt,ermany, Art 7734). All anhydrous reactions were carried outnder a nitrogen atmosphere. Solutions were dried over anhy-rous sodium sulfate.

Mesitylene, sodium cyanide, lithium aluminum hydride,thyl bromopyruvate, chloroacetone, and 2-bromoacetophenoneere obtained from commercial chemical companies and usedithout further purification. Solvents were obtained and driedy usual laboratory techniques. Poly(vinyl chloride) (PVC)nd bis(2-ethylhexyl) adipate (DOA) were purchased fromluka Chemie AG (Buch, Switzerland). 1,3-Bis[tris(hydroxy-ethyl)methylamino]propane (Bis–tris propane) and 2-amino-

-(hydroxymethyl)-1,3-propanediol (Tris) were purchased fromigma (St. Louis, MO, USA). All other chemicals for ana-

ytical experiments were of analytical-reagent grade. Standard

olutions and buffers were prepared with freshly deionizedater (18 M� cm). 1,3,5-Tris(cyanomethyl)-2,4,6-trimethyl-enzene (1) was prepared by known literature procedure5].

tripodal thiazole derivatives.

1 nta 7

2

2(

2rra173δ

C5

2t

v4wT2E186915fC

22

(Teeec1R(N61

MC5

2t

(

ACroE1(21(CN

2t

arwcER(1

3N151

2e

datgwtFeftta(cstHdi

988 H.-S. Kim et al. / Tala

.2. Synthesis

.2.1. 1,3,5-Tris(thioamidomethyl)-2,4,6-trimethylbenzene2)

H2S gas was introduced to a mixture of 1 (500 mg,.11 mmol) and triethylamine (10 mL) in a DMF (30 mL) atoom temperature for 5 h with stirring. After the solvent wasemoved, the residue was crystallized with CH2Cl2 and hex-ne to give 2 (550 mg, 77%). TLC, Rf 0.40 (100% EtOAc); mp95–196 ◦C; IR (KBr) 3273, 3133, 1632, 1425, 1215, 951, 789,21 cm−1; 1H NMR, δ: 8.99 (s, 3H, NH), 7.45 (s, 3H, NH),.61 (s, 6H, ArCH2CSNH2), 1.65 (s, 9H, ArCH3); 13C NMR,: 204.6 (CSNH2), 135.8, 130.2, 46.4, 15.7; Anal. Calcd. for15H21N3S3: C, 53.06; H, 6.23; N, 12.38; S, 28.33; Found: C,3.21; H, 6.03; N, 12.16; S, 28.16.

.2.2. 1,3,5-Tris[2′(4′-carbethoxythiazolyl)methyl]-2,4,6-rimethylbenzene (3)

A mixture of 2 (1.0 g, 2.95 mmol) and ethyl bromopyru-ate (1.9 g, 9.74 mmol) in dry EtOH (50 mL) was refluxed forh. After the solvent was removed, the mixture was extractedith CH2Cl2. The organic layer was dried and concentrated.he residue was chromatographed (elution with EtOAc:hexane,:1) on silica gel to give 3 (1.69 g, 91%). TLC, Rf 0.50 (100%tOAc); mp 219◦C (CH2Cl2–hexane); IR (KBr) 3097, 2982,719, 1480, 1335, 1230, 1099, 1022, 777 cm−1; 1H NMR, δ:.01 (s, 3H, ThzH), 4.54 (s, 6H, ArCH2Thz), 4.44 (q, J = 6.9 Hz,H, ThzCO2CH2CH3), 2.29 (s, 9H, ArCH3), 1.42 (t, J = 6.9 Hz,H, ThzCO2CH2CH3); 13C NMR, δ: 172.1, 161.3, 147.2, 136.5,33.9, 127.1, 61.4, 34.8, 17.0, 14.2; MS, m/z: 627 (M+, 100),94 (45), 512 (29), 494 (29), 470 (65), 300 (33); Anal. Calcd.or C30H33N3O6S3: C, 57.39; H, 5.30; N, 6.69; S, 15.32; Found:, 57.59; H, 5.34; N, 6.55; S, 15.43.

.2.3. 1,3,5-Tris[2′(4′-hydroxymethylthiazolyl)methyl]-,4,6-trimethylbenzene (4)

LiAlH4 (115 mg, 3.01 mmol) was added to a solution of 3317 mg, 0.50 mmol) in a dry THF (20 mL) in an ice bath.he mixture was stirred for 6 h at the same temperature andthyl acetate was added to the resulting mixture to destroyxcess LiAlH4. After the solvent was removed, the residue wasxtracted with ethyl acetate. The organic layer was dried andoncentrated. The residue was chromatographed (elution with00% ethyl acetate) on silica gel to give 4 (190 mg, 75%). TLC,f 0.21 (1:9, CH3OH:EtOAc); mp 184 ◦C (CH3OH–H2O); IR

KBr) 3420, 2920, 1628, 1464, 1385, 1127, 1026 cm−1; 1HMR, δ: 7.08 (s, 3H, ThzH), 5.26 (bs, 3H, ThzCH2OH), 4.65 (s,H, ThzCH2OH), 4.41 (s, 6H, ArCH2Thz), 2.29 (s, 9H, ArCH3);3C NMR, δ: 170.5, 157.3, 135.4, 133.6, 113.4, 60.0, 34.2, 16.6;

S, m/z: 501 (M+, 100), 355 (26), 258 (30); Anal. Calcd. for24H27N3O3S3: C, 57.46; H, 5.42; N, 8.38; S, 19.17; Found: C,7.39; H, 5.68; N, 7.98; S, 18.96.

.2.4. 1,3,5-Tris[2′(4′-methylthiazolyl)methyl]-2,4,6-rimethylbenzene (5)

A solution of 2 (214 mg, 0.63 mmol) and chloroacetone292 mg, 3.16 mmol) in benzene (20 mL) was refluxed for 6 h.

meNo

1 (2007) 1986–1992

fter the solvent was removed, the residue was extracted withH2Cl2. The organic layer was dried and concentrated. The

esidue was chromatographed (elution with EtOAc:hexane, 2:1)n silica gel to give 5 (135 mg, 47%). TLC, Rf 0.28 (1:1,tOAc:hexane); mp 138 ◦C (CH2Cl2–hexane); IR (KBr) 2919,526, 1459, 1305, 1163, 1120, 719, 611 cm−1; 1H NMR, δ: 6.67s, 3H, ThzH), 4.43 (s, 6H, ArCH2Thz), 2.43 (s, 9H, ThzCH3),.30 (s, 9H, ArCH3); 13C NMR, δ: 170.7, 152.6, 136.0, 134.0,12.9, 34.7, 17.1, 16.9; MS, m/z: 453 (M+, 100), 396 (22), 34135), 242 (50), 229 (48), 112 (47); Anal. Calcd. for C24H27N3S3:, 63.54; H, 6.00; N, 9.26; S, 21.20; Found: C, 63.16; H, 6.16;, 8.95; S, 21.12.

.2.5. 1,3,5-Tris[2′(4′-phenylthiazolyl)methyl]-2,4,6-rimethylbenzene (6)

A solution of 2 (204 mg, 0.60 mmol) and 2-bromo-cetophenone (598 mg, 3.00 mmol) in benzene (20 mL) wasefluxed for 6 h. After the solvent was removed, the residueas extracted with CH2Cl2. The organic layer was dried and

oncentrated. The residue was chromatographed (elution withtOAc:hexane, 1:1) on silica gel to give 6 (161 mg, 42%). TLC,f 0.56 (1:2, EtOAc:hexane); mp 102 ◦C (CH2Cl2-hexane); IR

KBr) 3064, 2918, 1601, 1489, 1296, 1148, 1071, 733 cm−1;H NMR, δ: 7.87–7.90 (m, 6H), 7.32–7.44 (m, 9H), 7.30 (s,H, ThzH), 4.56 (s, 6H, ArCH2Thz), 2.42 (s, 9H, ArCH3); 13CMR, δ: 171.2, 155.5, 136.2, 134.6, 134.1, 128.7, 128.0, 126.3,12.4, 34.9, 17.1; Anal. Calcd. for C39H33N3S3: C, 73.20; H,.20; N, 6.57; S, 15.03; Found: C, 72.98; H, 5.13; N, 6.51; S,5.20.

.3. Preparation of electrodes and their potentiometricvaluation

Ion-selective membrane cocktails [33,34] were prepared byissolving a newly synthesized ionophore 1 wt%, PVC 33 wt%nd DOA 66 wt% in a THF (1 mL). The cocktail solutions werehen poured into a glass ring (i.d. 22 mm) placed on a slidelass, and dried at room temperature for a day. Small disksere punched from the cast films and mounted in Phillips elec-

rode bodies (IS-561; Glasblaserei Moller, Zurich, Switzerland).or all electrodes, 0.1 M of KCl was used as the internal refer-nce electrolyte. All electrodes were presoaked in distilled wateror 1 h before use. Potential differences between the ISEs andhe Orion sleeve-type double junction Ag/AgCl reference elec-rode (Model 90-02) were measured using a PC equipped withhigh-impedance input 16-channel analog-to-digital converter

KOSENTECH Inc., Busan, Korea). The dynamic responseurves were obtained at room temperature by adding standardolutions to 200 mL of a magnetically stirred background elec-rolyte (0.05 M of Tris–HCl, pH 7.4; 0.05 M of Bis–tris propan-

2SO4, pH 9.0; 0.01 M of magnesium acetate–HNO3, pH 4.5;eionized water) every 100 s to vary the concentration of eachonic species stepwise from 10−6 to 10−1 M. The potentials were

easured every second at room temperature. The response of thelectrodes to pH changes was tested by adding aliquots of a 1NaOH solution to a solution of 11.4 mM of boric acid, 6.7 mMf citric acid, and 10.0 mM of NaH2PO4 at room temperature.

nta 71 (2007) 1986–1992 1989

Sscsa

3

eimiipci3rtptrvaSerd

tctTiFt

Fiv

F((

tmtapetabatamp

H.-S. Kim et al. / Tala

electivity coefficients were estimated according to the separateolution-matched potential method (IUPAC SSM II method) byomparing the activity of an interfering cation that induced theame potential change as that which was induced by ammoniumctivity of 1.0 × 10−2 M [35].

. Results and discussion

Newly synthesized ion-selective neutral carriers 3–6 werevaluated with DOA-plasticized PVC membranes. In measur-ng their potentiometric responses to various cations, e.g., alkalietal, alkaline earth metal cations, and ammonium ion, the ligat-

ng properties of the tripodal binding site (nitrogens and sulfursn a pendant thiazole ring, or pendant carbonyl, hydroxyl andhenyl groups) was studied. Since the interaction between aation and the thiazole nitrogen may compete with the protonn a solution, we examined the potentiometric responses of the-, 4-, 5-, and 6-based membrane electrodes to varying pH. Theesults are summarized in Fig. 2. The compound 5-based elec-rode exhibited the most sensitive responses to protons with aH between 3 and 8. Other electrodes show the pH responses inhe same range, but less sensitive than the 5-based one. The pHesponse test suggests that the proton affinity of compounds 3–6aries with the type of substituent used on pendant thiazoles; theyre in the order of –CH3, –COOC2H5 ≈ –C6H6 and –CH2OH.ince the pH responses of the 3-, 4-, 5-, and 6-based membranelectrodes begin to level down near pH 8, we examined theiresponses to other cations under near neutral and alkaline con-itions.

At pH of 7.4 (0.05 M of Tris–HCl), 4-, 5- and 6-based elec-rodes exhibited negligible responses to common physiologicalations (Li+, Na+, K+, Ca2+ and Mg2+) and a slightly increasedo ammonium (�E change of 50–80 mV from 10−3 to 10−1 M).

he 3-based electrode, on the other hand, exhibited substantially

ncreased responses to ammonium and similarly to potassium.ig. 3 shows the slopes of 47.2 and 45.0 mV per decade and

he detection limits of 1.3 × 10−4 and 5.0 × 10−4 M, respec-

ig. 2. The pH responses of the 3-, 4-, 5-, and 6-based membrane (1 wt.% ofonophore in 33 wt.% of PVC plasticized with 66 wt.% of DOA) electrodes toarying pH levels.

frtae5atbicciTrtcts6mch

ig. 3. The potentiometric responses of the 3-based membrane electrode to Li+

-�-), Na+ (-�-), K+ (-�-), NH4+ (-©-), Mg2

+ (-�-), and Ca2+ (-�-) at pH 7.4

0.05 M Tris–HCl).

ively. Considering the similar responses of the 3- and 6-basedembranes to protons, the noticeably increased responses of

he 3-based membrane, compared to those of the 6-based one tommonium and potassium, may suggest that the carbonyl grouparticipates in the ligation of those cations. A simple, semi-mpirical modeling using MOPAC at the AM1 level confirmshat three thiazole nitrogens and carbonyl oxygens form a hex-dentate cavity that can hold a cation [32]. The average distancesetween the cation and thiazole nitrogens and between the cationnd carbonyl oxygens are about 3.1 and 2.9 A, respectively. Onhe other hand, the small potentiometric responses of the 4-, 5-nd 6-based electrodes, to all cations at near neutral pH levelsay indicate that thiazole nitrogens are partially screened by

rotons.Also, we examined the same potentiometric responses of the

our electrodes at pH 9 (0.05 M Bis–tris propan-H2SO4); theesults are shown in Fig. 4. The elevated pH greatly increasedhe potentiometric responses to ammonium and potassium ofll electrodes. Only the 3-based membrane electrode, how-ver, exhibited near Nernstian responses (58.9 mV/pNH4

+ and7.0 mV/pK+) to both ammonium and potassium, with a rel-tively low detection limit (2 × 10−6 and 5 × 10−6 M, respec-ively). The nearly identical potentiometric response of the 3-ased membrane electrode, to both ammonium and potassium,mplicate that the binding site of compound 3 is a size-selectiveavity. It seems that the electrostatic interaction between theavity and a size-fitting cation contributes more to the selectiv-ty of the binding site than does hydrogen-bonding interaction.he 4- and 5-based membrane electrodes exhibit less sensitive

esponses to ammonium and potassium than the 3-based elec-rode, while showing similar potentiometric responses to otherations. The results implicate that the cation-binding ability ofhe tripodal thiazoles is similar to each other otherwise the sub-tituents on thiazoles are capable of capping the binding site. The

-based membrane electrode exhibited much inferior potentio-etric responses to the other three electrodes; the semi-empirical

alculation model suggests that the bulky phenyl substituentsinder the formation of a symmetrical binding site. The poten-

1990 H.-S. Kim et al. / Talanta 71 (2007) 1986–1992

F repar( -), an

te

bteAbamvtpea

t

tsetcietirbli

TT

C

ig. 4. The potentiometric responses of the DOA-plasticized PVC membranes p6), containing thiazole: Li+ (-�-), Na+ (-�-), K+ (-�-), NH4

+ (-©-), Mg2+ (-�

iometric characteristics of the 3-, 4-, 5-, and 6-based membranelectrodes are summarized in Table 1.

Sulfur atoms in tripodal thiazole derivatives may provide ainding site for the transition metal ions. To examine the transi-ion metal ion selectivity of the 3-, 4-, 5-, and 6-based membranelectrodes, the potentiometric responses of the four electrodes tog+, Fe2+, Co2+, Zn2+, Ni2+, Pb2+, Cd2+, Cu2+ and Hg2+ haveeen measured under acidic conditions (0.01 M of magnesiumcetate–HNO3, pH 4.5). Acidic conditions prevent the transitionetal ions from forming precipitates with hydroxide, and it pro-

ides favorable silver ion binding ability [36]. It was interestingo observe that all four electrodes exhibited nearly negligibleotentiometric responses to all transition metal cations with the

xception of Ag+: the linear dynamic ranges were between 10−5

nd 10−3 M.To examine the response characteristics of the four electrodes

o Ag+, without the effect of interferents, measurements were

dwbi

able 1he potentiometric properties of the thiazole derivatives in DOA-plasticized PVC me

ompound Slopea Detection limitb Selectivity

K+

3 58.9 −5.3 −0.24 46.8 −4.0 −0.25 56.2 −4.1 −0.26 34.9 −3.9 −0.2

a Slopes from 10−4 to 10−1 M (mV/decade).b log [NH4

+].

ed with the tripodal carbethoxy (3), hydroxymethyl (4), methyl (5), and phenyld Ca2

+ (-�-).

aken in deionized water. The results are shown in Fig. 5, andummarized in Table 2. The response characteristics of the fourlectrodes to silver ion do not show a clear correlation withheir responses to ammonium. The compound with methyl groupontaining thiazole, 5, showed a Nernstian response to Ag+

n the 1 × 10−8 to 1 × 10−4 M range. On the other hand, thelectrode based on 4, which showed similar response charac-eristics to those based on 5, had the poorest response to silveron (35.9 mV/pAg+ in the 1 × 10−7 to 3 × 10−3 M range). Theesponse of the 6-based electrode was similar to that of the 4-ased one. The semi-empirical modeling study (PM3) providesittle information regarding the role of the thiazole substituentsn silver binding, while there are some changes in the electron

ensity levels on sulfur atoms, depending on the substituents,e were not inclined to draw conclusions at this time. The 3-ased electrode exhibited near a Nernstian response to silvern the 3 × 10−8 to 1 × 10−3 M range. The molecular modeling

mbranes

coefficient(

log kpot

NH+4 , j

)

Na+ Li+ Ca2+ Mg2+

−2.5 −3.1 −4.0 −4.5−1.0 −1.2 −2.6 −3.4−1.1 −1.3 −2.7 −3.2−0.8 −1.0 −1.0 −1.6

H.-S. Kim et al. / Talanta 71 (2007) 1986–1992 1991

Fig. 5. The potentiometric responses of the electrodes to Ag+ based on neutral carriers 3, 4, 5 and 6: (left) dynamic response curves; (right) calibration curves.

Table 2The potentiometric response properties of the 3-, 4-, 5-, and 6-based membrane electrodes

No. Ionophore Slopea (mV/decade) Detection limitb, −log [Ag+/M] Linear range (M)

1 Compound 3 51.3 −8.4 3 × 10−8 to 1 × 10−3

2 Compound 4 35.9 −7.9 1 × 10−7 to 3 × 10−3

3 Compound 5 59.5 −8.6 1 × 10−8 to 1 × 10−4

4 −7 −3

sbi

4

mavscritlttfOeewi3(lNtfbdn

A

veS

R

[

[[

[[[[[

Compound 6 39.3

a Slopes from the linear range.b Logarithmic scale.

tudy suggests that the high potentiometric responses of the 3-ased electrode to silver ion could be attributed to the long-rangenteraction of the substituents on thiazole.

. Conclusion

The potentiometric responses of the 3-, 4-, 5-, and 6-basedembrane electrodes to alkali metal, alkaline earth metal,

mmonium, and transition metal ions have been examined underarious pH conditions. The 3-based electrode exhibited a Nern-tian response to ammonium and potassium under alkaline pHonditions. The other three electrodes showed poor potentiomet-ic performances in both near neutral and high pH conditions,mplicating that the binding ability of the cavity, formed byhree thiazoles, is not sufficient to hold alkali metal or alka-ine earth metal cations including ammonium. It was suggestedhat the carbethoxy group, which contains thiazole deriva-ive 3, forms a size-selective binding site, which in turn wasormed by three thiazole nitrogens and three carbonyl groups.n the other hand, the potentiometric responses of the four

lectrodes to transition metal cations were quite different; alllectrodes showed substantial responses to silver ion, but thereas almost no response to all other transition metal ions exam-

ned (Fe2+, Co2+, Zn2+, Ni2+, Pb2+, Cd2+, Cu2+ and Hg2+). The- and 5-based electrodes resulted in near Nernstian responses51.3 mV and 59.5 mV/pAg+, respectively) with low detectionimits (∼100 ppt), while the 4- and 6-based ones showed sub-ernstian responses below 40 mV/pAg+. It was assumed that

he difference in the potentiometric responses to silver ion arose

rom the difference in the charge density of sulfur atoms inducedy a long range, through the bond interaction of substituents. Aetailed theoretical study is in progress in order to examine theature of such interactions.

[

[

−8.0 3 × 10 to 1 × 10

cknowledgements

This work was supported by the Kyungpook National Uni-ersity Research Team Fund, 2003. HN also gratefully acknowl-dges the support of the Basic Research Program of the Koreacience & Engineering Foundation (R01-2006-000-10240-0).

eferences

[1] J.W. Steed, J.L. Atwood, Supramolecular Chemistry, John Wiley & SonsLtd., 2000.

[2] K. Wygladacz, A. Radu, C. Xu, Y. Qin, E. Bakker, Anal. Chem. 77 (2005)4706.

[3] X.-B. Zhang, Z.-X. Han, Z.-H. Fang, G.-L. Shen, R.-Q. Yu, Anal. Chim.Acta 562 (2006) 210.

[4] D.D. MacNicol, A.D.U. Hardy, D.R. Wilson, Nature 266 (1976) 611.[5] G. Hennrich, E.V. Anslyn, Chem. Eur. J. 8 (2002) 2219.[6] S. Sasaki, D. Citterio, S. Ozawa, K. Suzuki, J. Chem. Soc. Perkin Trans. 2

(2001) 2309.[7] H.J. Choi, Y.S. Park, S.H. Yun, H.-S. Kim, C.S. Cho, K. Ko, K.H. Ahn,

Org. Lett. 4 (2002) 795.[8] D.R. Turner, M.J. Paterson, J.W. Steed, J. Org. Chem. 71 (2006) 1598.[9] J. Kim, B. Raman, K.H. Ahn, J. Org. Chem. 71 (2006) 38.10] H.R. Seong, D.-S. Kim, S.G. Kim, H.J. Choi, K.H. Ahn, Tetrahedron Lett.

45 (2004) 723.11] S.-G. Kim, K.H. Ahn, Chem. Eur. J. 6 (2000) 3399.12] A. Vacca, C. Nativi, M. Cacciarini, R. Pergoli, S. Roelens, J. Am. Chem.

Soc. 126 (2004) 16456.13] T.D.P. Stack, Z. Hou, K.N. Raymond, J. Am. Chem. Soc. 115 (1993) 6466.14] C.M. Hartshorn, P.J. Steel, Chem. Commun. (1997) 541.15] R.H. Holm, P. Kennepohl, E.I. Solomon, Chem. Rev. 96 (1996) 2239.16] T.D.P. Stack, R.H. Holm, J. Am. Chem. Soc. 109 (1987) 2546.17] C. Walsdorff, W. Saak, S. Pohl, J. Chem. Soc., Dalton Trans. (1997)

1857.18] C. Walsdorff, S. Park, J. Kim, J. Heo, K.-M. Park, J. Oh, K.J. Kim, Chem.

Soc., Dalton Trans. 6 (1999) 923.19] J. Chin, C. Walsdorff, B. Stanix, J. Oh, H.J. Oh, S.-M. Park, K. Kim, Angew.

Chem. Int. Ed. 38 (1999) 2756.

1 nta 7

[

[

[

[

[

[[

[

[

[

[

[

[[

992 H.-S. Kim et al. / Tala

20] H. Higuchi, T. Matsufuji, T. Oshima, K. Ohto, K. Inoue, T. Tsend-Ayush,K. Gloe, Chem. Lett. 34 (2004) 80.

21] X. Zeng, L. Weng, L. Chen, F. Xu, Q. Li, X. Leng, X. He, Z.-Z. Zhang,Tetrahedron 58 (2002) 2647.

22] L. Chen, H. Ju, X. Zeng, X. He, Z. Zhang, Anal. Chim. Acta 437 (2001)191.

23] M.J. Goldcamp, K. Ashley, S.E. Edison, J. Pretty, J. Shumaker, Electro-analysis 17 (2005) 1015.

24] I. Isildak, M. Yolcu, O. Isildak, G. Topal, H. Horsgoren, Microchim. Acta144 (2004) 177.

25] C.C. Su, M.C. Chang, L.L.K. Liu, Anal. Chim. Acta 432 (2001) 261.26] B.H. Kim, H.P. Hong, K.T. Cho, J.H. On, Y.M. Jun, I.S. Jeong, G.S. Cha,

H. Nam, Talanta 66 (2005) 794.27] J.H. Choi, Y.K. Koh, I.-C. Kwon, H.-S. Kim, H.J. Park, S.J. Kim, G.S. Cha,

H. Nam, Bull. Korean Chem. Soc. 20 (1999) 581.

[[

[

1 (2007) 1986–1992

28] H.-S. Kim, J.H. Choi, Y.K. Koh, H.J. Choi, G.S. Cha, H. Nam, J. InclusionPhenom. 40 (2001) 265.

29] H.-S. Kim, H.J. Park, H.J. Oh, Y.K. Koh, J.H. Choi, D.-H. Lee, G.S. Cha,H. Nam, Anal. Chem. 72 (2000) 4683.

30] J.H. Shim, I.S. Jeong, M.H. Lee, H.P. Hong, J.H. On, K.S. Kim, H.-S. Kim,B.H. Kim, G.S. Cha, H. Nam, Talanta 63 (2004) 61.

31] H.-S. Kim, K.S. Do, K.S. Kim, J.H. Shim, G.S. Cha, H. Nam, Bull. KoreanChem. Soc. 25 (2005) 1465.

32] P. Buhlmann, E. Pretsch, E. Bakker, Chem. Rev. 98 (1998) 1593.33] G. Craggs, G.J. Moody, J.D.R. Thomas, J. Chem. Educ. 51 (1974)

541.34] G.J. Moody, R.B. Oke, J.D.R. Thomas, Analyst 95 (1970) 910.35] Y. Umezawa, K. Umezawa, H. Sato, Pure Appl. Chem. 67 (1995)

507.36] J. Lu, X. Tong, X. He, J. Electroanal. Chem. 540 (2003) 111.