Selective recognition of fluoride and acetate by a newly designed ruthenium framework: experimental and theoretical investigations

DaltonTransactions

Dynamic Article Links

Cite this: Dalton Trans., 2012, 41, 4484

www.rsc.org/dalton PAPER

Selective recognition of fluoride and acetate by a newly designed rutheniumframework: experimental and theoretical investigations†

Tanaya Kundu,a Abhishek Dutta Chowdhury,a Dipanwita De,a Shaikh M. Mobin,a Vedavati G. Puranik,b

Anindya Datta*a and Goutam Kumar Lahiri*a

Received 7th November 2011, Accepted 13th January 2012DOI: 10.1039/c2dt12126c

An effective anion sensor, [RuII(bpy)2(H2L−)]+ (1+), based on a redox and photoactive {RuII(bpy)2}

moiety and a new ligand (H3L = 5-(1H-benzo[d]imidazol-2-yl)-1H-imidazole-4-carboxylic acid), hasbeen developed for selective recognition of fluoride (F−) and acetate (OAc−) ions. Crystal structures ofthe free ligand, H3L and [1](ClO4) reveal the existence of strong intramolecular and intermolecularhydrogen bonding interactions. The structure of [1](ClO4) shows that the benzimidazole N–H of H2L

− ishydrogen bonded with the pendant carboxylate oxygen while the imidazole N–H remains free forpossible hydrogen bonding interaction with the anions. The potential anion sensing features of 1+ havebeen studied by different experimental and theoretical (DFT) investigations using a wide variety ofanions, such as F−, Cl−, Br−, I−, HSO4

−, H2PO4−, OAc− and SCN−. Cyclic voltammetry and differential

pulse voltammetry established that 1+ is an excellent electrochemical sensor for the selective recognitionof F− and OAc− anions. 1+ is also found to be a selective colorimetric sensor for F− or OAc− anionswhere the MLCT band of the receptor at 498 nm is red shifted to 538 nm in the presence of oneequivalent of F− or OAc− with a distinct change in colour from reddish-orange to pink. The bindingconstant between 1+ and F− or OAc− has been determined to be logK = 7.61 or 7.88, respectively, basedon spectrophotometric titration in CH3CN. The quenching of the emission band of 1+ at 716 nm (λex =440 nm, Φ = 0.01 at 298 K in CH3CN) in the presence of one equivalent of F− or OAc−, as well as twodistinct lifetimes of the quenched and unquenched forms of the receptor 1+, makes it also a suitablefluorescence-based sensor. All the above experiments, in combination with 1H NMR, suggest theformation of a 1 : 1 adduct between the receptor (1+) and the anion (F− or OAc−). The formation of 1 : 1adduct {[1+·F−] or [1+·OAc−]} has been further evidenced by in situ ESI-MS(+) in CH3CN. Though thereceptor, 1+, is comprised of two N–H protons associated with the coordinated H2L

− ligand, only the freeimidazole N–H proton participates in the hydrogen bonding interactions with the incoming anions, whilethe intramolecularly hydrogen bonded benzimidazole N–H proton remains intact as evidenced by thecrystal structure of the final product (1). The hydrogen bond mediated anion sensing mechanism, over thedirect deprotonation pathway, in 1+ has been further justified by a DFT study and subsequent NBOanalysis.

Introduction

The design of receptors for the selective recognition of anions isone of the most important areas of current chemical research.1

This is primarily due to the fact that over 70% of co-factors andsubstrates in biology are anionic in nature and anions play

essential roles in the activity of enzymes, the transport of hor-mones, protein synthesis and DNA regulation along with variouschemical and environmental processes.2 For example, the oxy-anion, acetate, is a critical component in various metabolic pro-cesses and the rate of the production of acetate via oxidation hasbeen considered to be an indicator for organic decomposition inmarine sediments.3 Similarly, fluoride ions have a significantrole in dental care and for the treatment of osteoporosis.4 Apartfrom their essential roles, the presence of excessive fluoride andoxy-anions may cause diseases such as fluorosis and cysticfibrosis, primarily due to drinking water contamination.2a,5

Therefore, the development of suitable receptors and sensors forselective anions is a formidable challenge. Anions, being nega-tively charged, can bind either with the positively charged recep-tor through an electrostatic interaction or with the neutral

†Electronic supplementary information (ESI) available: characterizationdetails of the complexes, crystallographic material and DFT results(Tables S1–S9 and Fig. S1–S22). CCDC 827492, 827493 and 839295.For crystallographic data in CIF or other electronic format see DOI:10.1039/c2dt12126c

aDepartment of Chemistry, Indian Institute of Technology Bombay,Powai, Mumbai 400076, India. E-mail: [email protected] for Materials Characterization, National Chemical Laboratory,Dr. Homi Bhabha Road, Pune 411008, India

4484 | Dalton Trans., 2012, 41, 4484–4496 This journal is © The Royal Society of Chemistry 2012

Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online / Journal Homepage / Table of Contents for this issue

http://dx.doi.org/10.1039/c2dt12126c


www.rsc.org/dalton


http://pubs.rsc.org/en/journals/journal/DT

http://pubs.rsc.org/en/journals/journal/DT?issueid=DT041015

receptor through a hydrogen bonding interaction.6 Amide,pyrrole, urea and thiourea-based neutral anion receptors, as wellas guanidinium, quaternary ammonium, and different Lewis acid(tin, boron, silicon, mercury etc.)-based cationic anion receptorsare well documented.6 In recent years transition metal complexderived sensors have also been studied extensively, mainly dueto their long lifetime compared to the general organic receptors.7

Among them, ruthenium(II)-polypyridyl based complexes arefound to be a prominent class of receptor primarily because oftheir intense metal-to-ligand charge transfer (MLCT) transitionsin the visible region, which indeed facilitate the recognition oftheir anion binding feature simply by monitoring the change incolour.8 The luminescence and redox properties of these com-plexes are also affected greatly during the anion sensing process.

In this regard, the present article describes the design of [Ru(bpy)2(H2L

−)](ClO4) [1](ClO4), which incorporates the singlydeprotonated and newly synthesized H3L (H3L = 5-(1H-benzo[d]imidazol-2-yl)-1H-imidazole-4-carboxylic acid) and its anionsensing behaviour. H3L exists in a zwitterionic form in the solidstate.

Interestingly, 1+ is found to be an efficient receptor towardsthe selective recognition of fluoride and acetate ions.

Herein we report the synthesis, characterization and singlecrystal X-ray structures of H3L, [1](ClO4) as well as the deproto-nated product, 1. The potential application of 1+ as a suitablereceptor for the anions, F−, Cl−, Br−, I−, HSO4

−, H2PO4−, OAc−

and SCN− has been explored via experimental (electrochemical

and various spectroscopic techniques) and theoretical (DFT)investigations.

Results and discussion

Synthesis and characterization

The ligand, H3L, has been prepared by the reaction of a1 : 1 mixture of 4,5-imidazoledicarboxylic acid and o-phenyle-nediamine in viscous o-phosphoric acid. The complex, [RuII-(bpy)2(H2L

−)](ClO4) [1](ClO4), has been synthesized from a1 : 1.2 mixture of the in situ generated cis-[RuII(bpy)2(EtOH)2]

2+

and H3L under reflux in ethanol and a nitrogen atmosphere, fol-lowed by chromatographic purification using a neutral aluminacolumn (see Experimental section for details).

The free ligand, H3L in CH3OH, and [1](ClO4) in CH3CNexhibit molecular ion peaks at m/z = 229.09 and 641.16, respect-ively, corresponding to the calculated molecular masses of229.07 {H3L+H}

+ and 641.09 {[1]-ClO4}+ (Fig. S1†). H3L and

[1](ClO4) give satisfactory microanalytical data and [1](ClO4)exhibits 1 : 1 electrical conductivity (see Experimental section).

Crystal Structures of H3L and [1](ClO4). The crystal struc-tures of H3L and [1](ClO4) are shown in Fig. 1 and 2, respect-ively. The crystallographic parameters are shown in Table 1.Selected bond lengths and bond angles of H3L and [1](ClO4) arepresented in Table S1† and Table 2, respectively, which matchfairly well with the data obtained from the DFT optimised struc-tures (Fig. S2†). The ligand, H3L, has been crystallised withthree molecules of water. In the crystal of H3L the hydrogenatom of the carboxylic acid group is transferred from O1 to N3,the nitrogen atom of the benzimidazole moiety (Fig. 1),leading to a zwitterionic molecule with an intramolecular N3–H3⋯O1 hydrogen bond with N3–H3 and H3⋯O1 distances of0.88 Å and 1.89 Å, respectively (Table S2†). The packingdiagram of H3L shows the presence of intermolecular hydrogenbonds and the formation of water tetramers (Fig. S3a andTable S2†). The water molecules, O111 and O333, from two

Fig. 1 ORTEP diagram of H3L·3H2O. Ellipsoids are drawn at a 50% probability level.

This journal is © The Royal Society of Chemistry 2012 Dalton Trans., 2012, 41, 4484–4496 | 4485

Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


neighbouring layers generate a water tetramer. These water tetra-mers are further stabilised by intermolecular hydrogen bondingfrom the neighbouring molecules through N4–H4⋯O111 andO333–H444⋯O1 (carboxylate group of H3L) interactions.The isolated water molecule, O222, undergoes intermolecularhydrogen bonding with the N2 atom, forming an O222–

H666⋯N2 interaction. The other hydrogen atom of O222 formsa hydrogen bond with the water cluster through an O222–H555⋯O333 interaction.

The crystal structure of [1](ClO4) reveals that the unsymmetri-cal H2L

− is selectively coordinated to the {Ru(bpy)2}2+ unit

through the anionic carboxylate oxygen (O1) and the neutral

Fig. 2 ORTEP diagram of the cationic part of [1](ClO4). Ellipsoids are drawn at a 40% probability level. The perchlorate ion and solvents of crystal-lisation are removed for clarity.

Table 1 Crystallographic data and refinement parameters for H3L, [1](ClO4) and 1

Compound H3L·3H2O [1](ClO4)·2H2O 2[1]·5CH3CN

Empirical formula C11H14N4O5 C31H27ClN8O8Ru C72H59N21O4Ru2Mr 282.26 776.13 1484.54Crystal size/mm 0.23 × 0.16 × 0.09 0.33 × 0.29 × 0.23 0.34 × 0.26 × 0.22Crystal system Monoclinic Monoclinic MonoclinicSpace group P21/n C2/c P21A/Å 6.6996(5) 19.8792(7) 11.7216(3)b/Å 16.5546(8) 20.3466(5) 22.8856(7)c/Å 11.5646(11) 18.3949(6) 12.4104(4)α (°) 90 90 90β (°) 106.310(8) 117.034(4) 94.881(3)γ (°) 90 90 90V/Å3 1231.00(16) 6627.3(4) 3317.09(17)Z 4 8 2F(000) 592 3152 1516Flack value — — 0.27(2)μ/mm−1 0.122 0.617 0.524T/K 150(2) 150(2) 150(2)hkl range −7 to 7, −19 to 19, −13 to 13 −19 to 23, −24 to 23, −21 to 21 −13 to 13, −27 to 27, −13 to 14ρcalcd/g cm−3 1.523 1.556 1.486θ range (°) 3.40 to 25.00 3.33 to 25.00 3.41 to 24.99Reflns collected 8437 24 219 24 270Unique reflns (Rint) 2148 [0.0526] 5819 [0.0432] 11 620 [0.0367]Data/restraints/parameters 2148/0/205 5819/10/472 11 620/1/906R1 (I > 2σ(I)) 0.0344 0.0540 0.0350wR2 (all data) 0.0656 0.1530 0.0896GOF 0.830 0.876 1.013Largest diff. peak, hole/eÅ−3 0.157/−0.248 0.735/−0.383 0.776/−0.446


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


nitrogen donor (N1) of the imidazole moiety, forming a fivemember chelate ring with a bite angle of 79.99(18)° (Fig. 2 andTable 2). The average Ru–N(bpy) distance of 2.042 Å matchesreasonably well with the reported distances in analogous com-plexes.9 The Ru–N1 and Ru–O1 distances involving the coordi-nated ligand (H2L

−) of 2.049(5) Å and 2.102(4) Å, respectively,are slightly shorter with respect to the Ru–N(imidazole) and Ru–O(carboxylate) distances of 2.077/2.069 Å and 2.137 Å reportedin analogous Ru(bpy)2 complexes of 4,5-bis(benzimidazol-2-yl)imidazole and imidazole-4,5-dicarboxylic acid.10 The bite anglesinvolving H2L

− (N(1)–Ru–O(1)) and bpy (N(6)–Ru–N(5)/N(8)–Ru–N(7)) of 79.99(18)° and 78.7(2)°/78.96(17)°, respectively,and trans-angles, N(6)–Ru–N(1)/N(7)–Ru–N(5)/N(8)–Ru–O(1)of 167.25(17)°/176.2(2)°/172.56(17)° collectively imply a dis-torted octahedral geometry around the ruthenium(II) ion in [1](ClO4).

The packing diagram displays intra- and inter-molecular N–H⋯O hydrogen bonding interactions in [1](ClO4). The intramo-lecular hydrogen bonding interaction involves the benzimidazolehydrogen atom, N3–H3 and the uncoordinated carboxylateoxygen atom (O2), where the H3⋯O2 distance is 1.88 Å(Table S3 and Fig. S3b†).

Anion sensing aspects of 1+

Electrochemistry. 1+ exhibits quasi-reversible one-electronoxidation at E298° = 0.69 V and two successive one-electronreductions at −1.66 and −1.93 V versus SCE in CH3CN(Fig. S4†). The DFT calculations at the B3LYP level suggest thatthe HOMO or HOMO-1 are dominated by the ligand (H2L

−)-based orbitals (97% or 99%) and an appreciable metal contri-bution (72%) has only been predicted at the lower HOMO-2

level, which suggests a ligand-centred preferential oxidationprocess (Table 3 and Fig. S5†). Thus, the otherwise expectedmetal-based (RuII/RuIII) oxidation process is silent in the presentcase due to the redox non-innocent feature of the coordinatedH2L

− in 1+.11 On the other hand, the expected stepwise bpy-based reductions are supported by the observation that theLUMO or LUMO+1 are dominated by bpy-based orbitals (93%)(Table 3 and Fig. S5†). The ligand (H2L

−)-based oxidationprocess is further evidenced by the free radical type EPR signal

Table 2 Selected bond lengths (Å) and angles (°) for [1](ClO4)

Bond lengths Bond angles

X-Ray DFT X-Ray DFT

Ru–N(5) 2.060(5) 2.100 N(6)–Ru–N(8) 96.01(17) 94.33Ru–N(6) 2.031(5) 2.089 N(6)–Ru–N(7) 97.54(18) 97.86Ru–N(7) 2.041(4) 2.086 N(8)–Ru–N(7) 78.96(17) 78.05Ru–N(8) 2.037(4) 2.097 N(8)–Ru–N(1) 95.73(18) 97.84Ru–N(1) 2.049(5) 2.121 N(7)–Ru–N(1) 89.58(16) 88.88Ru–O(1) 2.102(4) 2.108 N(6)–Ru–N(5) 78.7(2) 78.20N(1)–C(3) 1.245(7) 1.323 N(1)–Ru–N(5) 94.2(2) 95.19N(1)–C(2) 1.420(6) 1.388 N(6)–Ru–O(1) 88.85(17) 89.50N(2)–C(4) 1.327(7) 1.378 N(7)–Ru–O(1) 94.86(16) 94.08N(2)–C(3) 1.398(7) 1.357 N(1)–Ru–O(1) 79.99(18) 79.00N(3)–C(5) 1.340(6) 1.368 N(5)–Ru–O(1) 85.32(15) 86.20N(3)–C(11) 1.406(7) 1.378 N(8)–Ru–N(5) 101.13(18) 101.05N(4)–C(5) 1.307(7) 1.323 N(6)–Ru–N(1) 167.25(17) 167.09N(4)–C(6) 1.387(7) 1.382 N(7)–Ru–N(5) 176.2(2) 175.91C(6)–C(11) 1.394(8) 1.423 N(8)–Ru–O(1) 172.56(17) 171.64C(1)–C(2) 1.460(8) 1.492 O(1)–C(1)–O(2) 124.3(6) 125.01C(2)–C(4) 1.358(7) 1.385C(4)–C(5) 1.503(8) 1.442C(1)–O(1) 1.256(7) 1.301C(1)–O(2) 1.271(7) 1.236C(6)–C(7) 1.407(8) 1.404C(7)–C(8) 1.384(9) 1.387C(8)–C(9) 1.396(9) 1.414C(9)–C(10) 1.366(9) 1.389C(10)–C(11) 1.367(8) 1.399

Table 3 Molecular orbital compositions of 1+

M.O. Energy (eV)

Composition

Ru bpy H2L

LUMO+9 −2.404 0.45 0.29 0.26LUMO+8 −2.458 0.13 0.33 0.53LUMO+7 −2.518 0.05 0.70 0.25LUMO+6 −3.333 0.01 0.05 0.94LUMO+5 −3.660 0.03 0.96 0.02LUMO+4 −3.779 0.03 0.95 0.02LUMO+3 −3.807 0.04 0.95 0.01LUMO+2 −4.062 0.03 0.95 0.02LUMO+1 −4.748 0.06 0.93 0.01LUMO −4.902 0.06 0.93 0.01HOMO −7.516 0.02 0.01 0.97HOMO−1 −7.690 0.01 0.00 0.99HOMO−2 −7.986 0.72 0.11 0.17HOMO−3 −8.076 0.72 0.15 0.13HOMO−4 −8.328 0.78 0.14 0.08HOMO−5 −8.982 0.03 0.02 0.95HOMO−6 −9.363 0.02 0.01 0.97HOMO−7 −9.411 0.00 0.00 1.00HOMO−8 −9.625 0.02 0.78 0.20HOMO−9 −9.772 0.13 0.29 0.58


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


of the coulometrically generated oxidised species (12+) at 110 Kwith g = 1.999 (Fig. S6a†). Though the spin-density plot of 12+

(Fig. S6b†) shows Mulliken spin densities on H2L− and Ru of

0.333 and 0.412, respectively, the expected metal-based aniso-tropy has not been resolved in the EPR spectrum of the oxidisedspecies.

Cyclic voltammetry (CV) and differential pulse voltammetry(DPV) have been utilised to monitor the anion sensing ability of1+ in CH3CN. The gradual addition of anions (F− or OAc−) hasled to a sequential decrease in the anodic current of the oxidationprocess (E298° = 0.69 V) with the concomitant appearance of anew couple at a relatively low potential (E298° = 0.47 V) (Fig. 3)without any significant change in potential of the bpy-basedreduction processes. On addition of one equivalent of the F− orOAc− anion, the initial oxidation couple of 1+ at 0.69 V disap-pears completely with the simultaneous growth of a new coupleat 0.47 V (Fig. 3). The large negative shift in the potential(>200 mV) on addition of the anion in 1+ indicates the involve-ment of a strong hydrogen bonding interaction between the N2–H2 proton of the coordinated H2L

− of 1+ and the anion, F− orOAc−, leading to the eventual formation of 1 via the abstractionof H2, which gives HF or HOAc, respectively (Scheme 1 andFig. S7–S10†).12 The negative shift in the oxidation potential onaddition of the anion originates due to the decrease in theHOMO–LUMO energy gap to 1.65 eV in 1 from 2.61 eV in 1+.The formation of 1 (Scheme 1) via the interaction with anions(F− or OAc−) and 1+ in a 1 : 1 ratio has also been observed inthe single crystal X-ray structure (Table 1 and Fig. S11,Table S4†).

Unlike F− or OAc−, in the presence of one equivalent ofH2PO4

−, only partial transformation of 1+ to 1 takes place asindicated by the existence of both the couples at 0.69 V and 0.47V, which correspond to 1+ and 1, respectively, implying a weakinteraction between 1+ and H2PO4

−. However, in the presence ofother anions (Cl−, Br−, I−, HSO4

− and SCN−, one-equivalent ormore), no such changes in the CV and DPV are noticed, whichin effect suggests that there is no interaction or a very weak inter-action between 1+ and the said anions. The above observationsthus establish that 1+ is an excellent selective electrochemicalsensor for fluoride (F−) and acetate (OAc−) anions.

Absorption spectroscopy. Besides the intense UV region tran-sition in CH3CN, 1

+ exhibits one moderately intense transitionin the visible region at 498 nm with a shoulder in the higherenergy region at 446 nm (see Experimental section). The visibleand UV region bands are assigned to the (dπ)RuII → (π*)bpyand intra-ligand π–π* transitions based on the TD–DFT calcu-lations on the optimised structure of 1+ (Fig. S2b andTable S5†).

The anion binding feature of 1+ has also been investigatedspectrophotometrically in acetonitrile. The addition of an aceto-nitrile solution of the TBA (tetrabutylammonium) salt of Cl−,Br−, I−, HSO4

− or SCN− up to eight equivalents, with regards tothe acetonitrile solution of 1+, does not alter the electronic spec-trum of 1+ (Fig. S12†). The lowest energy MLCT band of 1+ at498 nm has, however, been gradually red-shifted to 538 nm bythe addition of one equivalent of F− or OAc− (Fig. 4) and thenewly developed band at 538 nm has been assigned to theRuII(dπ) → bpy(π*) MLCT transition according to the TD–DFT

calculations on the optimised structure of A or B in Scheme 1(Tables S6–S9† and Fig. 4). The appreciable red-shift (40 nm)of the MLCT band upon addition of the anion (F− or OAc−) canbe rationalised on the basis of a decrease in the HOMO–LUMOenergy gap to 1.82 eV in A or B (Scheme 1) with respect to 1+,where the energy gap is 2.61 eV. This in turn suggests that astrong hydrogen bonding interaction is operational between theF− or OAc− anion and the N2–H2 proton of H2L

− in 1+, whicheventually increases the electron density on the coordinatedH2L

− ligand.

Fig. 3 Changes in the cyclic voltammogram (the differential pulse vol-tammogram is shown in the inset) of [1](ClO4) (10−3 mol dm−3) inCH3CN upon gradual additions of (a) [TBA][F] and (b) [TBA][OAc] upto one equivalent.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


The transformation of 1+ to 1 in the presence of F− or OAc−

proceeds through several isosbestic points (Fig. 4), revealing theexclusive presence of 1+ and [1+⋯F−] (A) or [1+⋯OAc−] (B)during the course of the transformation process (Scheme 1 andFig. S7 and S9†). Furthermore, a distinct change in colour of thesolution from reddish-orange to pink takes place during thetransformation (Fig. 5). The addition of greater than one equival-ent of F− or OAc− in the above solution of 1+ does not changethe intensity as well as the position of the 538 nm band of 1,suggesting the formation of a 1 : 1 adduct between 1+ and F− orOAc− (A or B in Scheme 1). This 1 : 1 binding profile betweenthe receptor 1+ and F− or OAc− has also been reflected (inset ofFig. 4) in the simultaneous changes in the absorbance at 498 and538 nm as a function of the equivalents of the F− or OAc−

anion.

½RuðbpyÞ2ðLH2Þ�þ ð1þÞ þ X� ðF� or OAc�ÞO ½½RuðbpyÞ2ðLH2Þ�þ � X��! ½RuðbpyÞ2ðLHÞ�ð1Þ þ HF or HOAc ð1Þ

K ¼ ½1þ: x��½1þ� ½x��

ΔA

¼Δεð½S�þ½L�þ1=KÞ+ Δε2ð½S�þ½L�þ1=KÞ2�4Δε2½S�½L�

� �1=2

2ð2Þ

The binding constant, K, of 1+ and F− or OAc− has been cal-culated based on eqn (1) and (2).12a,b,13

In eqn (2), ΔA represents the change in the initial absorbanceof 1+ at 498 nm upon each addition of the anion F− or OAc−

and [S] and [L] are the concentrations of 1+ and F− or OAc−,respectively, during the spectrophotometric titrations. Thebinding constant, K, and the change in the molar extinctioncoefficient (Δε) are estimated from the nonlinear curve fittingprocedure using ΔA at each concentration of F− or OAc−

(Fig. S13†). The nonlinear curve fitting procedure, performedaccording to eqn (2), results in binding constant values of logK= 7.61 or 7.88 for F− and OAc−, respectively (Fig. S13†). Thecalculated high logK values imply a strong binding interactionbetween the receptor, 1+, and the anions, F− or OAc−.

On the other hand, under identical conditions, three equiva-lents of H2PO4

− are required for the complete transformation of1+ to 1 (Fig. S14†), implying a relatively weak interactionbetween the receptor, 1+, and H2PO4

− compared to F− or OAc−

as has been established earlier via the electrochemical studies.The said transformation of 1+ to 1 has been further established

via controlled spectrophotometric titrations of 1+ with a strongbase, TBAOH, in a 1 : 1 molar ratio (Fig. S15†).

The pKa1 value of the receptor, 1+, has been determined to be8.1 by monitoring the pH dependent spectral changes in 1 : 1CH3CN : H2O. However, the pKa2 for 1+ could not be deter-mined due to the precipitation problem at higher pH values evenat an experimental concentration of 5 × 10−5 mol dm−3. Theselective anion sensing feature of 1+ for F− can be attributed tothe order of the basicity of the halides: F− > Cl− > Br− > I−

(pKa(aq) values are 3.45, −7, −9 and −11 for HF, HCl, HBr andHI, respectively).14 Similarly, the stronger interaction of 1+ withOAc−, as compared to other oxy-anions and SCN−, can berationalized in terms of their pKa1(aq) values: HOAc (4.75) >H3PO4 (2.12) > H2SO4 (−2)12a,b,14 and SCN− (∼ − 2).15

However, it should to be pointed out that the collective effects,due to the basicity of the anions and the strength of hydrogenbonding interaction (see later), essentially control the transfer ofthe proton from the receptor (1+) to the anions.12b

Scheme 1 Formation of 1 via the abstraction of the H2 proton by F− or OAc−.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


1H and 19F NMR spectra. The 1H NMR spectrum of 1+ inCD3CN exhibits the calculated number of twenty-one partiallyoverlapping aromatic proton signals and two NH signals (seeExperimental section). The similar chemical shift of many ofthe aromatic protons has prevented the identification of the

individual proton signals. However, three singlets correspondingto two N–H protons and one C–H proton of the imidazole unitof the coordinated H2L

− ligand have been identified at 14.54,7.56 and 7.46 ppm, respectively (Fig. S16a†). The peak pos-itions of the two N–H protons have been authenticated by a D2Oexchange experiment (Fig. S16b†). In the presence of one equiv-alent of F− or OAc− the spectral pattern involving the aromaticprotons changes slightly. The N3–H3 (benzimidazole ring) andC–H (imidazole ring) proton signals of H2L

− in 1+ at 14.54 ppmand 7.46 ppm have been shifted to 14.90/14.95 ppm and6.92/6.66 ppm, respectively, on addition of one equivalent of F−/OAc−. However, the NH proton signal of the imidazole ring(N2–H2) of H2L

− in 1+ at 7.56 ppm disappears in the presenceof one equivalent of F− or OAc− (Fig. S17†), which implies theeventual abstraction of the H2(N2) proton as shown inScheme 1. This in turn increases the electron density in the imi-dazole ring, which pushes the imidazole C–H signal slightly upfield.

No noticeable change in the 1H NMR spectral profile hasbeen observed in the presence of excesses qualities of F− orOAc−, indicating that a 1 : 1 binding situation between the recep-tor, 1+, and the anion, which has also been established by elec-trochemistry and spectrophotometry.

The 19F NMR spectrum of TBAF in CD3CN exhibits oneintense signal at −117.93 ppm and another very weak signal at−150.96 ppm for TBAF and HF2

−, respectively.16 HF2− has

been formed due to the presence of slight moisture in the system.Upon addition of 0.6 equivalent of 1+ to the above TBAF sol-ution, the 19F signal at −117.93 ppm vanishes completely andthe intensity of the HF2

− signal at −150.90 ppm is increased sig-nificantly (Fig. S18†), which implies a hydrogen bonding inter-action between the N2–H2 proton of 1+ and the F− of TBAF,followed by an abstraction of the proton, which leads to the for-mation of HF2

− (HF + F− ⇌ HF2−).12

Emission and time resolved fluorescence. 1+ exhibits an emis-sion band at 716 nm (quantum yield (Φ) of 0.01) at 298 K inacetonitrile. The emission band at 716 nm has, however, beenquenched significantly by the addition of the F− or OAc− ion,along with a red-shift of the band to 760 nm (Fig. 6). On theother hand, no appreciable quenching of the emission band(716 nm) has been noticed in the presence of other anions, suchas Cl−, Br−, I−, HSO4

− or SCN− (Fig. S19†). This is in agree-ment with the selective binding of 1+ with F− and OAc− ions ashas been established by the other experimental techniques (elec-trochemistry, spectrophotometry and NMR). Furthermore, upon

Fig. 4 The changes in the absorption spectrum of 1+ (5 × 10−5 moldm−3) in CH3CN upon gradual additions of 0–1 equivalent of (a) [TBA][F] and (b) [TBA][OAc]. The insets show the changes in the absor-bances at 498 and 538 nm as a function of the equivalents of the anions,F− and OAc−.

Fig. 5 The visual change in colour of 1+ in CH3CN (5 × 10−5 mol dm−3) upon addition of one equivalent of the TBA salt of the anions, F−, Cl−,Br−, I−, H2PO4

−, HSO4− or OAc−.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


excitation at 440 nm, the quenching process has been found tocontinue to some extent even after addition of one equivalent ofF− or OAc− (inset of Fig. 6). This clearly indicates that, inaddition to ground state binding, excited state interactionsbetween the receptor, 1+, and the anions are also operative. Inorder to prove this, Stern–Volmer plots are shown in Fig. 7,where the upward curvature of the plot suggests quenching byboth ground state complex formation and excited state collisions.In other words, both static and dynamic quenching processes areoperative in the course of the binding interaction between 1+ andthe anion (F− or OAc−). On the other hand, in the presence ofH2PO4

−, the quenching process continues up to six equivalentsof the H2PO4

− anion (Fig. S20†), which confirms the selectiveformation of a 1 : 1 complex between the receptor, 1+, and theanion, F− or OAc−. It may be noted that the fluorescence is notquenched completely even at the highest concentration of theanions (10 equivalents of F− or OAc−). Rather, a red shiftedspectrum with λmax = 760 nm emerges with higher equivalent of

the anions (F− or OAc−). This spectrum is attributed to thedeprotonated form, 1, which is eventually formed via the inter-mediate hydrogen bonded species, A or B in Scheme 1. Therelationship between the emission band, λmax = 760 nm and thedeprotonated form, 1, has been authenticated by the developmentof an identical emission spectral pattern on addition of oneequivalent of TBAOH to 1+ (Fig. S21†).

The fluorescence decay of 1+ in the presence of F− and OAc−

are recorded at excitation and emission wavelengths of 440 nmand 716 nm, respectively (Fig. 8). Initially, a single componentof 31–35 ns exists while, upon addition of 0.6 equivalents of F−

and OAc−, a second component of 9–16 ns begins to appear(Table 4). Upon the gradual addition of the anion, the contri-bution of the longer component (A2, Table 4) decreases with theconcomitant increase of the shorter component (A1, Table 4).The longer lifetime is attributed to the unquenched form of thereceptor (1+), which when quenched (eventually becoming 1,Scheme 1) exhibits a shorter lifetime. Thus, the fluorescence

Fig. 6 Quenching of the emission intensity of 1+ upon addition of 0 to 10 equivalents of (a) TBAF and (b) TBAOAc in CH3CN at 298 K. Additionshave been made in increments of 0.1 equivalents from 0 to 1 equivalent. The numbers of equivalents for subsequent additions are 1.2, 1.5, 2.0, 5.0and 10 equivalents. The changes in the quantum yield (logΦ) as a function of the number of equivalents of the anions (F− and OAc−) are shown inthe inset.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


quenching of the receptor in the presence of selective anions andthe distinct lifetimes of the quenched and unquenched forms ofthe receptor, 1+, make it a suitable probe for fluorescence life-time sensing as well as the fluorescence intensity-based sensingof anions.

Theoretical (DFT) insights into the anion binding mode ofreceptor 1+. The presence of the intramolecular N3–H3⋯O2 hydrogen bond in the crystal of 1+ has also been sup-ported by DFT calculations (Fig. 2 and Fig. S2b, S3b†). Thecoordinated H2L

− ligand in 1+ has two N–H protons, which canin principle participate in the anion binding process throughhydrogen bonding interactions. The receptor (1+) interacts withF− and OAc− in a 1 : 1 molar ratio and greater than one equival-ent of F− or OAc− does not alter the electrochemical or the spec-troscopic signatures (see above). It is therefore logical toconsider that the intramolecular hydrogen bond H3(N3) is notparticipating in the anion binding process, which finds furtherjustification through DFT calculations (Fig. S7 and S9†).

Fig. 7 Stern–Volmer plots for (a) F− and (b) OAc−.

Fig. 8 The fluorescence decay of 1+ in the presence of (a) F− and (b)OAc−, added in increments of 0.2 equivalents up to 1.2 equivalents.

Table 4 The temporal parameters for the binding of F− and OAc−

with the receptor, 1+a

Equivalence of F− τ1 (ns) τ2 (ns) A1 A2 χ2

0.0 35 1.310.2 35 1.230.4 35 1.290.6 15 35 0.11 0.89 1.340.8 15 35 0.16 0.84 1.191.0 16 35 0.30 0.70 1.241.2 15 34 0.55 0.45 1.10Equivalence of OAc− τ1 (ns) τ2 (ns) A1 A2 χ2

0.0 33 1.190.2 33 1.170.4 32 1.210.6 9 32 0.11 0.89 1.150.8 12 32 0.19 0.81 1.201.0 12 32 0.35 0.65 1.151.2 12 31 0.52 0.48 1.13

a The decays are fitted to a biexponential function, I(t) = I(0)[A1 exp(−t/τ1) + A2 exp(−t/τ2)], where τ1 and τ2 are the two lifetimes and A1 and A2are the corresponding amplitudes. χ2 is the measure of the goodness ofthe fit.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


In principle, there are two possible binding modes of thereceptor, 1+, with the anion, F− or OAc−, either involving boththe N–H protons, N2–H2 (free) and N3–H3 (hydrogen bonded:N3–H3⋯O2) or through N2–H2 only (Scheme 2). In the case ofF−, binding mode A (Scheme 2) is calculated to be energeticallymore stable by 7 kcal mol−1 than the alternate mode A′ and, forOAc−, binding mode B (Scheme 2) is energetically more stableby 9 kcal mol−1 than B′. Interestingly, DFT calculations alsopredict that the N3–H3⋯O2 interaction in 1+ remains unaffectedin both these cases upon interaction with the anions (A and B inScheme 2) as has also been suggested by the experimentalstudies.

Though there are two possible mechanisms that can controlthe anion sensing process: (i) through hydrogen bonding inter-actions and (ii) through direct deprotonation,17 the presence of astrong hydrogen bond in A or B can, however, eliminate thepossibility of a direct deprotonation pathway. The hydrogenbond-mediated anion sensing process in the present case can berationalised on the basis of NBO calculations (Table 5). Thelonger N2–H2 distance in A (1.672 Å) as compared to that inoptimised 1+ (1.014 Å) and the strong bonding interactionbetween H2 and F− with a H2–F distance of 0.971 Å implies anintermediate situation where the N2–H2 bond is cleaved withconcomitant formation of a H2–F bond. A N2–H2–F angle of151.5° and D⋯A distance of 2.6 Å collectively suggest a hydro-gen bonded anion sensing mechanism. The decrease in the

natural charge on N2 and the simultaneous increase in thenatural charge on H2 in A (Table 5) with regards to optimised 1+

also support the hydrogen bonded anion sensing mechanism.The natural charge on nearby C3 and C4 of the imidazolemoiety significantly decreases in A (Scheme 2, Table 5) as aresult of the proton elimination, which is also reflected in theshortening of the N2–C3 and N2–C4 distances in A as comparedto 1+. This is due to the negative charge delocalisation on theimidazole ring containing C3–N2–C4.

A similar effect has also been observed in the case of acetatein B. The N2–H2 bond distance in B increases to 1.696 Å from1.014 Å in optimised 1+ with a simultaneous strong bondinginteraction of 1.01 Å between N2–H2 and the incoming OAc−

during the sensing process. A N2–H2–OAc angle of 167.3°,D⋯A distance of 2.71 Å and decrease and increase of thenatural charges on N2 and H1, respectively, in B are suggestiveof a similar hydrogen bonded anion sensing mechanism(Table 5).

It should be noted that the natural charges on H3 and O2 donot alter in A or B (Scheme 2) with respect to 1+ (Table 5) asexpected from the experimental observation that the N3–H3⋯O2 hydrogen bonding remains intact during the anionsensing process.

The slight increase of the natural charge on Ru in A or B withregard to 1+ (Table 5) arise primarily due to the change in theelectronic environment around the metal ion, which is also

Scheme 2 Two possible binding modes of the receptor, 1+, for F− (A and A′) and OAc− (B and B′).

Table 5 Summary of the NBO charges

Complex N2 H2 C3 C4 Ru O2 H3 F− OAc−

1+ −0.526 0.471 0.218 0.135 0.653 −0.671 0.470A −0.552 0.557 0.179 0.100 0.670 −0.680 0.461 −0.620B −0.553 0.523 0.182 0.101 0.674 −0.684 0.461 −0.7281 −0.485 0.179 0.079 0.673 −0.682 0.460


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


reflected in the red-shifted MLCT band in the electronic spec-trum during the anion sensing process.

The formation of A and B have also been confirmed fromtheir in situ mass (ESI+) spectra in CH3CN, which exhibit peaksat m/z = 660.096 and 700.173 corresponding to [1+·F−] (calcu-lated, 660.098) and [1+·OAc−] (calculated, 700.113), respect-ively (Fig. S22†). This is further justified by the crystal structuredetermination of the deprotonated form, 1, generated by theaddition of one equivalent of F− or OAc− in a CH3CN solutionof 1+ (Table 1 and Fig. S11, Table S4†).

In addition, the neutral pH of 1+ in deionised H2O signifiesthe virtual non-existence of the alternate equilibrium state: 1+

⇌1+ H+, which in turn establishes the hydrogen bonded adductformation (A or B in Scheme 2), as evidenced by ESI-MS asstated above.

Conclusion

An efficient ruthenium complex-derived anion sensor ([Ru(bpy)2(H2L

−)]+, 1+) has been developed for the selective recog-nition of fluoride and acetate anions. The selectivity of 1+

towards acetate and fluoride anions over other anions, such asCl−, Br−, I−, HSO4

−, H2PO4− and SCN− has been convincingly

established using a wide variety of experimental techniques. Thedetailed theoretical studies, in combination with experimentalobservations, reveal the formation of a hydrogen bonded 1 : 1adduct between the receptor and the anion, [1+·F−] or [1+·OAc−]followed by proton abstraction, which generates the deprotonatedform, 1.

Experimental

Materials

The starting complex, cis-[Ru(bpy)2Cl2]·2H2O, was preparedaccording to the reported procedure.18 Imidazole-4,5-dicar-boxylic acid and 1,2-phenylendiamine were purchased fromMerck, India. The tetrabutylammonium (TBA) salts of F−, Cl−,Br−, I−, HSO4

−, OAc−, H2PO4− and SCN− were purchased from

Sigma–Aldrich or Alfa Aesar. Other chemicals and solventswere reagent grade and used as received. For spectroscopic andelectrochemical studies, HPLC grade solvents were used.

Physical measurements

UV–vis and fluorescence spectra were recorded on Perkin–Elmer950 lamda spectrophotometer and Varian Cary Eclipse spectrofl-uorimeter, respectively. For the spectrophotometric and spectrofl-uorimetric titrations, 3 μL aliquots of the TBA salt of therespective anions (5 × 10−3 mol dm−3) in acetonitrile wereadded by a micro-syringe in each step in 3 cm3 of a 5 × 10−5

mol dm−3 acetronitrile solution of 1+ using a quartz cuvette witha 1 cm path length and a volume of 3 cm3. The quantum yield ofthe complex, 1+, was calculated with reference to a [Ru(bpy)3]

2+

standard (Φstd = 0.089 at 298 K in CH3CN at λex = 450 nm19).Time resolved fluorescence data were obtained using a pico-second pulsed diode laser (λex = 440 nm) based time-correlatedsingle photon counting (TCSPC) instrument from IBH (United

Kingdom) set at a magic angle. The FWHM of the instrumentresponse function at this wavelength was found to be 530 ps. Aresolution of 112 ps channel−1 was used. The decay traces, thusobtained, were fitted by the iterative reconvolution method,using IBH DAS 6.0 data analysis software, to single or biexpo-nential functions. Cyclic voltammetric, differential pulse voltam-metric and coulometric measurements were carried out using aPAR model 273A electrochemistry system. Platinum wireworking and auxiliary electrodes and an aqueous saturatedcalomel reference electrode (SCE) were used in a three-electrodeconfiguration. The supporting electrolyte was Et4NClO4 and thesolute concentration was 10−3 mol dm−3. The half-wave poten-tial, E298°, was set equal to 0.5(Epa + Epc), where Epa and Epc areanodic and cathodic cyclic voltammetric peak potentials, respect-ively. A platinum wire-gauze working electrode was used in thecoulometric experiments. For a typical titration, 20 μL aliquotsof the TBA salt of the respective anion (5 × 10−2 mol dm−3) inacetonitrile were added by a micropipette in each step in 10 cm3

of a 10−3 mol dm−3 acetronitrile solution of 1+. All experimentswere carried out under a nitrogen atmosphere. Elemental analysiswas carried out with a Perkin–Elmer 24 °C elemental analyser.Electrospray mass spectra were recorded on a Micromass Q-ToFmass spectrometer. Solution electrical conductivity was checkedusing a Systronic 304 conductivity bridge. The EPR measure-ments were made with a Varian model 109C E-line X-band spec-trometer fitted with a quartz dewar for 110 K. 1H NMR and 19FNMR spectra were obtained with a 400 MHz Bruker FT spec-trometer. Trifluoro-toluene was used as an internal standard inCD3CN for recording the 19F NMR spectra at 298K.

Synthesis

Synthesis of 5-(1H-benzo[d]imidazol-2-yl)-1H-imidazole-4-car-boxylic acid (H3L). A mixture of 4,5-imidazoledicarboxylic acid(1.56 g, 10 mmol) and o-phenylenediamine (1.08 g, 10 mmol)was added in viscous o-phosphoric acid (25 cm3) and it washeated, first, at 220 °C for 2 h and then at 250 °C for another2 h. The colour of the solution was finally changed to blue. Thesolution was cooled to room temperature and then poured intocrushed ice in a 500 cm3 beaker under vigorous stirring, whichresulted in a blue precipitate. The precipitate was filtered off andthen added to water (100 cm3) before being slowly treated with25% aqueous ammonia until the colour changed to light pink(pH ≈ 8). The solid mass thus obtained was filtered and washedthoroughly with distilled water. This dried material was dis-solved in a minimum quantity of hot N,N-dimethylformamide(DMF) and treated with decolourizing carbon. After hot fil-tration, the filtrate was added to water in small portions with stir-ring, and the white solid product was filtered off and dried undervacuum. Yield: 1.09 g (48%). Anal. Calcd for C11H8N4O2: C,57.88; H, 3.54; N, 24.56%. Found: C, 57.62; H, 3.48; N,24.42%. ESI-MS(+) (m/z, CH3OH): 229.09 (corresponding to{H3L+H}

+, Calcd 229.07). 1H NMR (400 MHz, (CD3)2NCOD):δ(ppm): 18.36 (s, 1H, NH), 13.71 (s, 1H, NH), 8.19 (s,1H), 8.02(s, 4H).

Synthesis of [(bpy)2RuII(H2L

−)](ClO4), [1](ClO4). The start-ing complex, cis-[Ru(bpy)2Cl2]·2H2O, (100 mg, 0.20 mmol) andAgClO4 (108.6 mg, 0.52 mmol) were added in 15 cm3 absolute


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


ethanol and the mixture was heated at reflux for 2 h with stirring.The initial violet solution changed to orange–red. The mixturewas then cooled and filtered through a Gooch sintered glassfunnel. The ligand, H3L, (55 mg, 0.24 mmol) was added to theabove solution containing [Ru(bpy)2(EtOH)2]

2+. The resultingmixture was heated to reflux for 12 h under a nitrogen atmos-phere. The resultant solution was reduced to 5 cm3 and kept at0 °C overnight. The reddish precipitate, which formed oncooling, was filtered and washed thoroughly with ice-cold waterfollowed by cold ethanol and diethyl ether. The solid mass waspurified by column chromatography using a neutral aluminacolumn. The desired reddish–orange band was eluted withCH2Cl2–CH3CN (2 : 1). Upon removal of the solvent underreduced pressure the pure complex, [1](ClO4), was obtained as adark coloured solid, which was further dried under vacuum.Yield: 92 mg, 62%. Anal. Calcd for RuC31H23N8O6Cl: C 50.27,H 3.13, N 15.14%. Found: C, 50.08; H, 2.94; N, 14.88%. ΛM

(Ω−1 cm2 M−1, CH3CN, 298K): 94. ESI-MS(+) (m/z, CH3CN):641.16 (corresponding to {[1]-ClO4}

+, Calcd: 641.16). 1H NMR(400 MHz, CD3CN): δ, (ppm, J(Hz)): 14.54 (s, 1H, N3H3),8.92 (d, 1H, 5), 8.47 (q, 2H, 8, 5, 8), 8.38 (q, 2H, 8, 7, 8 Hz),8.27(d, 1H, 5), 8.04 (m, 2H), 7.88 (m, 2H), 7.83 (m, 1H), 7.70(d, 1H, 5), 7.64 (m, 2H), 7.56 (s, 1H, N2H2), 7.50 (m, 1H), 7.46(s, 1H, imidazole C-H), 7.24 (m, 4H), 7.18 (m, 1H). λ[nm] (ε[M−1cm−1], CH3CN): 498 (9838), 446 (6533), 354(15804), 324(25 114), 292 (66 461), 244 (29 424).

Caution. Perchlorate salts of metal complexes with organicligands are potentially explosive. Heating of the dried samplesmust be avoided and handling of small amounts has to proceedwith great caution.

Crystallography

Single crystals of H3L and [1](ClO4) were grown by slow evap-oration of their 1 : 1 N,N′-dimethylformamide–water and 1 : 1acetonitrile–acetone solutions, respectively. Single crystals of 1were prepared from the CH3CN solution of 1+ in the presence ofone equivalent of F− or OAc−. The crystal data were collectedon an Oxford X-CALIBUR-S CCD diffractometer at 150 K.Selected data collection parameters and other crystallographicresults are summarised in Table 1. All data were corrected for theLorentz polarisation and absorption effects. The programpackage of SHELX-97 was used for the structure solution andfull matrix least squares refinement on F2.20 Hydrogen atomswere included in the refinement using the riding model. H3L wascrystallised with three molecules of water. There are two ClO4

anions with half occupancy along with a cluster of four watermolecules, O9, O10, O11 and O12 in [1](ClO4)·2H2O. Thewater molecule O11 is in a general position, while the rest of thewater molecules are in special position with 1/4, 1/2 and 1/4occupancies for O9, O10 and O12, respectively, resulting in twowater molecules. The hydrogen atoms associated with the clusterof water molecules in [1](ClO4) could not be located. The contri-bution from these four hydrogen atoms was included in the cal-culation but not in the refinement. The asymmetric unit of 1 iscomprised of two complex molecules and five molecules ofCH3CN. The CCDC numbers of H3L, [1](ClO4) and 1 are827492, 827493 and 839295, respectively.

Computational details

Full geometry optimisations were carried out with tightthresholds at the (R)B3LYP and (U)B3LYP levels for 1+, A, B, 1and 12+ respectively, using density functional theory with Gaus-sian 03 (revision C.02).21 All the elements, except ruthenium,were assigned the 6-31G(d) basis set. The LanL2DZ basis setwith an effective core potential was employed for the rutheniumatom.22 Vertical electronic excitations based on the B3LYP opti-mised geometries were computed for the time-dependent densityfunctional theory (TD-DFT) formalism23 in acetonitrile usingthe Polarisable continuum model (PCM) of Tomasi and co-workers.24–26 The conductor-like PCM (CPCM), in conjugationwith the united atom topological model (using UAO radii,implemented in the Gaussian 03 (revision C.02)), was applied.The Gaussum package was used to calculate the fractional con-tributions of various groups to each molecular orbital.27 No sym-metry constraints were imposed during the structuraloptimisations and the nature of the optimised structures andenergy minima were defined by subsequent frequency calcu-lations. Natural bond orbital (NBO) analysis was performedusing the NBO 3.1 module of Gaussian 03 on the optimised geo-metries.28 All the calculated structures were visualised withChemCraft.29

Acknowledgements

Financial support received from the Department of Science andTechnology, Council of Scientific and Industrial Research (NewDelhi, India), is gratefully acknowledged. X-Ray structuralstudies were carried out at the National Single Crystal X-ray Dif-fraction Facility, Indian Institute Technology Bombay. Specialacknowledgment is made to the Sophisticated Analytical Instru-ment Facility (SAIF), Indian Institute of Technology, Bombay,for providing the EPR facility.

References

1 (a) Special Issue: Anion Coordination Chemistry II, Coord. Chem. Rev.,2006, 250 (b) V. Vajpayee, Y. H. Song, M. H. Lee, H. Kim, M. Wang, P.J. Stang and K.-W. Chi, Chem.–Eur. J., 2011, 17, 7837–7844; (c) P.A. Gale, Coord. Chem. Rev., 2001, 213, 79–128; (d) C. R. Bondy and S.J. Loeb, Coord. Chem. Rev., 2003, 240, 77–99; (e) P. D. Beer and P.A. Gale, Angew. Chem., Int. Ed., 2001, 40, 486–516; (f ) A. Chakraborty,R. Gunupuru, D. Maity, S. Patra, E. Suresh and P. Paul, Inorg. Chem.Commun., 2010, 13, 1522–1526; (g) A. Ghosh, B. Ganguly and A. Das,Inorg. Chem., 2007, 46, 9912–9918.

2 (a) P. Gamez, T. J. Mooibroek, S. J. Teat and J. Reedijk, Acc. Chem. Res.,2007, 40, 435–444; (b) M. Arunachalam, I. Ravikumar and P. Ghosh, J.Org. Chem., 2008, 73, 9144–9147; (c) R. Dutzler, E. B. Campbell,M. Cadene, B. T. Chait and R. MacKinnon, Nature, 2002, 415, 287–294;(d) H. Maeda and Y. Ito, Inorg. Chem., 2006, 45, 8205–8210.

3 (a) M. Arunachalam and P. Ghosh, Org. Lett., 2010, 12, 328–331;(b) F. Zapata, A. Caballero, A. Tárraga and P. Molina, J. Org. Chem.,2010, 75, 162–169; (c) X.-F. Shang and X.-F. Xu, BioSystems, 2009, 96,165–171; (d) T. Y. Joo, N. Sing, G. W. Lee and D. O. Jang, TetrahedronLett., 2007, 48, 8846–8850.

4 (a) T. W. Hudnall, C.-W. Chiu and F. P. Gabbai, Acc. Chem. Res., 2009,42, 388–397; (b) K. L. Kirk, Biochemistry of the Halogens and InorganicHalides, Plenum Press, New York, 1991, 58; (c) M. Kleerekoper, Endo-crinol. Metab. Clin. North Am., 1998, 27, 441–452; (d) E. J. Cho, J.W. Moon, S. W. Ko, J. Y. Lee, S. K. Kim, J. Yoon and K. C. Nam, J. Am.Chem. Soc., 2003, 125, 12376–12377; (e) Y. Bao, B. Liu, H. Wang,J. Tian and R. Bai, Chem. Commun., 2011, 47, 3957–3959.


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


5 (a) F. M. Ashcroft, Ion Channels and Disease, Academic Press, SanDiego, CA, 2000; (b) E. T. R. Guzmán, E. O. Regil, L. R. R. Gutiérrez,M. V. E. Alberich, A. R. Hernández and E. O. Regil, Contamination ofcorn growing areas due to intensive fertilization in the high plane ofMexico, Water, Air, Soil Pollut., 2006, 175, 77–98.

6 (a) R. Martínez-Máñez and F. Sancenón, Chem. Rev., 2003, 103, 4419–4476; (b) P. D. Beer, Acc. Chem. Res., 1998, 31, 71–80; (c) P. D. Beerand E. J. Hayes, Coord. Chem. Rev., 2003, 240, 167–189.

7 (a) S. J. Dickson, S. C. G. Biaginib and J. W. Steed, Chem. Commun.,2007, 4955–4957; (b) P. D. Beer, Chem. Commun., 1996, 689–696; (c) S.J. Dickson, M. J. Paterson, C. E. Willans, K. M. Anderson and J.W. Steed, Chem.–Eur. J., 2008, 14, 7296–7305; (d) J. G. Vos and J.M. Kelly, Dalton Trans., 2006, 4869–4883; (e) S. Derossi, H. Adams andM. D. Ward, Dalton Trans., 2007, 33–36.

8 (a) T.-P. Lin, C.-Y. Chen, Y.-S. Wen and S.-S. Sun, Inorg. Chem., 2007,46, 9201–9212; (b) M. H. V. Huynh, D. M. Dattelbaum and T. J. Meyer,Coord. Chem. Rev., 2005, 249, 457–483; (c) F. Szemes, D. Hesek,Z. Chen, S. W. Dent, M. G. B. Drew, A. J. Goulden, A. R. Graydon,A. Grieve, R. J. Mortimer, T. Wear, J. S. Weightman and P. D. Beer,Inorg. Chem., 1996, 35, 5868–5879; (d) P. D. Beer, S. W. Dent and T.J. Wearb, J. Chem. Soc., Dalton Trans., 1996, 2341–2346; (e) E. Berni,I. Gosse, D. Badocco, P. Pastore, N. Sojic and S. Pinet, Chem.–Eur. J.,2009, 15, 5145–5152.

9 (a) W. R. Browne, C. M. O’Connor, H. P. Hughes, R. Hage, O. Walter,M. Doering, J. F. Gallagher and J. G. Vos, J. Chem. Soc., Dalton Trans.,2002, 4048–4054; (b) W. R. Browne, D. Hesek, J. F. Gallagher, C.M. O’Connor, J. S. Killeen, F. Aoki, H. Ishida, Y. Inoue, C. Villani and J.G. Vos, Dalton Trans., 2003, 2597–2602; (c) D. P. Rillema, D. S. Jones,C. Woods and H. A. Levy, Inorg. Chem., 1992, 31, 2935–2938; (d) C.J. da Cunha, S. S. Fielder, D. V. Stynes, H. Masui, P. R. Auburn andA. B. P. Lever, Inorg. Chim. Acta, 1996, 242, 293–302; (e) B. K. Santra,M. Menon, C. K. Pal and G. K. Lahiri, J. Chem. Soc., Dalton Trans.,1997, 1387–1393.

10 (a) D. Saha, S. Das, C. Bhaumik, S. Dutta and S. Baitalik, Inorg. Chem.,2010, 49, 2334–2348; (b) S. Das, D. Saha, C. Bhaumik, S. Dutta andS. Baitalik, Dalton Trans., 2010, 39, 4162–4169.

11 M. M. Khusniyarov, E. Bill, T. Weyhermüller, E. Bothe andK. Wieghardt, Angew. Chem., Int. Ed., 2011, 50, 1652–1655.

12 (a) T. Kundu, S. M. Mobin and G. K. Lahiri, Dalton Trans., 2010, 39,4232–4242; (b) Y. Cui, H.-J. Mo, J.-C. Chen, Y.-L. Niu, Y.-R. Zhong,K.-C. Zheng and B.-H. Ye, Inorg. Chem., 2007, 46, 6427–6436;(c) Y. Cui, Y.-L. Niu, M.-L. Cao, K. Wang, H.-J. Mo, Y.-R. Zhong andB.-H. Ye, Inorg. Chem., 2008, 47, 5616–5624; (d) H.-J. Mo, Y.-L. Niu,M. Zhang, Z.-P. Qiao and B.-H. Ye, Dalton Trans., 2011, 40,8218–8225.

13 (a) Y. Liu, B.-H. Han and Y.-T. Chen, J. Phys. Chem. B, 2002, 106,4678–4687; (b) Y. Liu, B.-H. Han, S.-X. Sun, T. Wada and Y. Inoue,J. Org. Chem., 1999, 64, 1487–1493; (c) C. Basu, S. Biswas, A.P. Chattopadhyay, H. Stoeckli-Evans and S. Mukherjee, Eur. J. Inorg.Chem., 2008, 4927–4935; (d) L. Wang, X.-J. Zhu, W.-Y. Wong,J.-P. Guo, W.-K. Wong and Z.-Y. Li, Dalton Trans., 2005, 3235–3240;(e) B. Valeur, J. Pouget, J. Bouson, M. Kaschke and N. P. Ernsting,J. Phys. Chem., 1992, 96, 6545–6549; (f ) Y. Liu, B. Li, C.-C. You,T. Wada and Y. Inoue, J. Org. Chem., 2001, 66, 225–232.

14 D. F. Shriver and P. W. Atkins, Inorganic Chemistry, Oxford UniversityPress, 3rd edn, 1999.

15 (a) T. D. B. Morgan, G. Stedman and P. A. E. Whincup, J. Chem. Soc.,1965, 4813–4822; (b) T. I. Crowell and M. G. Hankins, J. Phys. Chem.,1969, 73, 1380–1383.

16 (a) C.-I. Lin, S. Selvi, J.-M. Fang, P.-T. Chou, C.-H. Lai andY.-M. Cheng, J. Org. Chem., 2007, 72, 3537–3542; (b) Z. Xu, N.J. Singh, S. K. Kim, D. R. Spring, K. S. Kim and J. Yoon, Chem.–Eur.J., 2011, 17, 1163–1170.

17 J. Wang, F.-Q. Bai, B.-H. Xia, L. Sun and H.-X. Zhang, J. Phys. Chem.A, 2011, 115, 1985–1991.

18 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17,3334–3341.

19 M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Sharda, A. J. Thomson,J. L. Glasper and D. J. Robbins, J. Chem. Soc., Perkin Trans. 2, 1984,1293–1301.

20 G. M. Sheldrick, SHELX-97, Program for Crystal Structure Solution andRefinement, University of Göttingen, Germany, 1997.

21 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,J. R. Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N. Kudin,J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A.D. Daniels, M. C. Strain, O. Farkas, D. K. Mazlick, A. D. Rabuck,K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaus-sian 03(Revision C.02), Gaussian Inc, Wallingford, CT, 2004.

22 (a) T. H. Dunning Jr. and P. J. Hay, in Modern Theoretical Chemistry, ed.H. F. Schaefer, Pleum, New York, 3rd edn, 1976, pp. 1–28; (b) P. J. Hayand W. R. Wadt, J. Chem. Phys., 1985, 82, 299–310.

23 (a) R. Bauernschmitt and R. Ahlrichs, Chem. Phys. Lett., 1996, 256,454–464; (b) R. E. Stratmann, G. E. Scuseria and M. J. Frisch,J. Chem. Phys., 1998, 109, 8218–8224; (c) M. E. Casida, C. Jamorski,K. C. Casida and D. R. Salahub, J. Chem. Phys., 1998, 108, 4439–4449.

24 S. Miertus, E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55, 117–129.25 M. Cossi, V. Barone, R. Cammi and J. Tomasi, Chem. Phys. Lett., 1996,

255, 327–335.26 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–2001.27 (a) N. M. O’Boyle, GaussSum 2.1, 2007, available at http://gausssum.sf.

net; (b) N. M. O’Boyle, A. L. Tenderholt and K. M. Langner, J. Comput.Chem., 2008, 29, 839–845.

28 (a) E. D. Glendening, A. E. Reed, J. E. Carpenter and F. Weinhold, NBO,version 3.1, Theoretical Chemistry Institute, University of Wisconsin,Madison, WI, 2001; (b) A. E. Reed and F. Weinhold, J. Chem. Phys.,1985, 83, 1736–1740; (c) A. E. Reed, L. A. Curtiss and F. Weinhold,Chem. Rev., 1988, 88, 899–926; (d) F. Weinhold, in Encyclopedia ofComputational Chemistry, ed. P. v. R. Schleyer, Wiley, New York, 1998,pp. 1792–1811.

29 D. A. Zhurko and G. A. Zhurko, ChemCraft 1.5, Plimus, San Diego, CA,92130, http://www.chemcraftprog.com


Dow

nloa

ded

by R

MIT

Uni

on

04 M

arch

201

3Pu

blis

hed

on 2

2 Fe

brua

ry 2

012

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2D

T12

126C

View Article Online


Documents

Selective recognition of fluoride and acetate by a newly designed ruthenium framework: experimental and theoretical investigations