PAPER www.rsc.org/dalton | Dalton Transactions

Complex formation between Ru(g6-p-cym)(H2O)3]2+ and (O,O) donor ligandswith biological relevance in aqueous solution†

Linda Bıro, Etelka Farkas and Peter Buglyo*

Received 13th May 2010, Accepted 2nd August 2010DOI: 10.1039/c0dt00469c

The interaction between [Ru(h6-p-cym)(H2O)3]2+ and (O,O) type chelators with different basicity of thedonor atoms was studied using combined pH-potentiometric, 1H-NMR and ESI-TOF-MS techniques.The studied nine ligands are building blocks of reported complexes with antitumor activity or maymodel (O,O) donor serum components capable of interacting with the administered half-sandwichruthenium(II) type drug. Composition and stability constants of the [Ru(h6-p-cym)(O,O)Y] type species(Y: H2O or OH-) were determined (T = 25.0 ◦C; I = 0.20 M (KCl)) and the metal ion binding strengthsof the ligands are discussed. It was found that ligands with two low basicity O donors (oxalic andcyclobutane-1,1-dicarboxylic acid) bind the metal ion at acidic conditions but are not able to preventthe hydrolysis at physiologically relevant conditions. Ligands with one low and one high basicity Odonor (lactic and salicylic acid) are weak binders for Ru(h6-p-cym)(H2O)3]2+. Pyrone or pyridinonebased ligands are capable of binding the metal ion over a wide pH range and no hydrolysis product isdetectable at pH = 7.4. The obtained speciation models may help in the rationalization of the biologicalactivity of such complexes and provide a deeper insight into the solution behaviour of half-sandwichRu(II) complexes with potential anticancer activity.

Introduction

Half-sandwich Ru(II) complexes with potential anticancer activityare attracting increasing attention. In these molecules amongthe three available co-ordination sites two are very often takenby a chelating ligand with (N,N) donor set.1 Less informa-tion is available about the more labile complexes in whichan (O,O) chelator is present. The Sadler group has studiedthe solution behaviour of [Ru(h6-p-cym)(acac)Cl] or [Ru(h6-p-cym)(trop)Cl] (in which p-cym is 1-Me-4-iPr-benzene, acac ispentane-2,4-dionate while trop is tropolonate (7-oxocyclohepta-1,3,5-trienolate); Chart 1) and determined the pKa value of theaqua species formed after the replacement of the chloride ligandby a water molecule.2,3 They have also found that at pH 2.2besides [Ru(h6-p-cym)(acac)(H2O)]+, [Ru(h6-p-cym)(H2O)3]2+ isco-present, while at pH > 9 a dinuclear mixed hydroxo complex,[{Ru(h6-p-cym)(acac)}2(m2-OH)]+, is suggested.2 Detailed NMRstudies by the same group on [Ru(h6-p-cym)(malt)Cl] (malt ismaltolate, Chart 1) showed that depending upon the concentrationof the complex at pH 7.4, [Ru(h6-p-cym)(malt)H2O]+ partiallyhydrolyzes in a fast process and [{Ru(h6-p-cym)}2(m2-OH)3]+ isalso formed. Moreover, malt capable of co-ordinating via a five-membered (O,O) chelate was found to prevent the hydrolysismore effectively than acac which forms a less stable six-memberedchelate.2,4 Recently Keppler et al. have studied the effect of thereplacement of the (O,O) donor malt with a (S,O) chelate set.The (S,O) analogue of malt was assumed to form a more stable

Department of Inorganic and Analytical Chemistry, University of Debrecen,H-4010, Debrecen, Hungary. E-mail: buglyo@delfin.unideb.hu† Electronic supplementary information (ESI) available: Fig. S1: depen-dence on pH of the 1H-NMR spectrum of a solution containing [Ru(h6-p-cym)Cl2]2 and Hmalt in a 1 : 1 ratio in D2O at 298 K. cdimer = 1.00 mM;Fig. S2: representative measured and simulated ESI-TOF-MS spectra ofthe various complexes. See DOI: 10.1039/c0dt00469c

complex than malt and, unlike [Ru(h6-p-cym)(malt)Cl], was shownto retain the anticancer activity of the complex in vivo.5 TheDyson group has found that [Ru(h6-p-cym)(ox)(pta)] or [Ru(h6-p-cym)(cbd)(pta)] (in which ox is oxalate, cbd is cyclobutane-1,1-dicarboxylate and pta is 1,3,5-triaza-7-phosphatricyclo[3.3.1.1]-decane; Chart 1) undergo similar hydrolytic reactions and thepresence of the (O,O) chelators provides different hydrolyticcharacter of the complexes.6,7 Recently the groups of Severinand Dyson studied the pH-dependent interconversion of themonomeric [Ru(h6-p-cym)(hp)(H2O)] (in which hp is a substituted3-hydroxy-2-pyridone acting as O,O donor; Chart 1) into atrinuclear complex in aqueous solution at elevated pH with theinvolvement of the pyridone-N in the co-ordination too.8

When dissolving in polar solvents and/or entering the bodyhalf-sandwich [Ru(h6-p-cym)(XY)(Z)] type complexes (XY = NN,NO or OO donor chelator; Z = monodentate ligand) undergoextensive transformation and may also interact with low as wellas high molecular mass components of the different biofluids.Thermodynamic and kinetic factors affecting these interactionsare extensively discussed in a recent review.9 The rate of dissoci-ation of the chelating ligand and the strength of the Ru(II)-(XY)interaction, and the pH range of the hydrolysis of the aqua speciesformed, seem to be key issues when metal-based anticancer drugswith in vivo activity are designed.1,10,11 Furthermore, as shown byBergamo and co-workers the interaction of these complexes withserum albumin remarkably changed their cytotoxicity in vitro.12

Similarly, Lay and co-workers have found that NAMI-A and otheroctahedral Ru(III) complexes form adducts with bovine serumalbumin that are important in understanding the anti-metastaticproperties of Ru complexes.13

Since the basicity of the different (O,O) chelators can be varied ina wide range this single parameter alone may allow one to fine tunethe stability of these types of complexes and control their activityat physiologically relevant conditions. Despite the above demands

10272 | Dalton Trans., 2010, 39, 10272–10278 This journal is © The Royal Society of Chemistry 2010

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

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

Chart 1

only very few detailed solution equilibrium studies on [Ru(h6-p-cym)(O,O)(H2O)] type compounds, exploring the full speciationof the system, are available in the literature.

Recently we have shown that one class of importantbiomolecules, monohydroxamates (also being (O,O) chela-tors), are capable of forming stable complexes with [Ru(h6-p-cym)(H2O)3]2+ both in the solid state and in aqueous solution.14 Wehave also explored in detail the hydrolysis of this Ru2+ species usingdifferent techniques and determined the stability constants ofthe hydroxo complexes formed in the H+–[Ru(h6-p-cym)(H2O)3]2+

system (see Experimental for data). Taking into consideration hy-drolysis, we were able to obtain the pH dependent speciation of theH+–[Ru(h6-p-cym)(H2O)3]2+–meaha system (meaha = N-methyl-acetohydroxamate) and determine the composition and stabilityconstants of the complexes formed with the monohydroxamate.14

In order to compare the [Ru(h6-p-cym)(H2O)3]2+ bindingstrength of meaha with that of other (O,O) donor ligands asystematic study has been carried out. These latter (O,O) chelatorsare (i) building blocks in ruthenium complexes with potential

anticancer activity, and (ii) they are biofluid components or modelsfor them. Ligands capable of forming five membered (ox, lact,koji, malt, dhp, tir) or six-membered (cbd, acac, sal) chelates,in which the donor atoms exhibit a variation of basicity, werechosen (Chart 1) and the interaction of these chelators with[Ru(h6-p-cym)(H2O)3]2+ was investigated. This paper summarisesthe results of a detailed solution equilibrium work in whichour goal was to estimate and to compare the composition andstability constants of the various Ru-(O,O) type complexes formedapplying three techniques: pH-potentiometry, 1H-NMR and ESI-TOF-MS.

Experimental

Materials and methods

RuCl3·xH2O, a-terpinene, acetylacetone, maltol, kojic acid, 1,2-dimethyl-3-hydroxy-4(1H)-pyridinone (dhp), cyclobutane-1,1-dicarboxylic acid, sodium oxalate, sodium lactate, salicylic

This journal is © The Royal Society of Chemistry 2010 Dalton Trans., 2010, 39, 10272–10278 | 10273

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online

acid, tiron, AgCF3SO3 were commercial products (Aldrich orFluka) and used as received. [Ru(h6-p-cym)Cl2]2 and [Ru(h6-p-cym)(acetone)3](CF3SO3)2 were synthesized according to literatureprocedures.15,16

Solution studies

For the solution studies doubly deionised and ultra filtered waterwas obtained from a Milli-Q RG (Millipore) water purificationsystem. pH-potentiometric measurements were carried out at anionic strength of 0.20 M (KCl) and at 25.0 ± 0.1 ◦C. Carbonate-freeKOH solutions of known concentrations (ca. 0.2 M) were used astitrant. HCl stock solutions were prepared from concentrated HCland their concentrations were determined by potentiometric titra-tions using the Gran’s method.17 A Mettler Toledo DL50 titratorequipped with a DG 114-SC combined electrode was used forthe pH-metric measurements. The electrode system was calibratedaccording to Irving et al.,18 the pH-metric readings could thereforebe converted into hydrogen ion concentration. The water ioniza-tion constant, pKw, is 13.76 ± 0.01 under the conditions employed.Titrations were performed in the pH range 2.0–11.0. Initial volumeof the samples was 15.00 mL. Metal ion stock solutions weremade by dissolving [Ru(h6-p-cym)(acetone)3](CF3SO3)2 in water.The ruthenium and proton concentrations were estimated with theaid of potentiometric titrations. The ligand concentrations werevaried in the range 0.002–0.004 M and the metal ion to ligandratios in the range 1 : 1–1 : 4. pH equilibrium was assumed to bereached if a change in the measured potential was less than 0.01 mVwithin 20 s. During the titrations, if a pH value was reached wherethe time was too long to obtain pH equilibrium (more than 15 minwaiting time) from this pH upward; the obtained data were omittedfrom the calculations. The reproducibility of the titration pointsincluded in the evaluation was within 0.005 pH unit. Typically 150–200 titration points were used for the calculations. The acceptedfitting of the titration curves was always less than 0.008 mL. Thesamples were in all cases completely deoxygenated by bubblingwith purified argon for ca. 20 min before the measurements. Thetitrations were carried out in a vessel equipped with normal jointsfor the electrode and for the base capillary. Concentration stabilityconstants, bp,q,r = [RupAqHr]/[Ru]p[A]q[H]r (where “Ru” stands for[Ru(h6-p-cym)(H2O)3]2+, “A” for the deprotonated form of theligands) were calculated with the aid of the PSEQUAD computerprogram.19 For the general formulae the charges are omitted ifthe different ligands have different charges. The following hydroxospecies and stability constants were taken into consideration inthe calculations: [{Ru(h6-p-cym)}2(m2-OH)2]2+ (log b2,0,-2: -7.12)and [{Ru(h6-p-cym)}2(m2-OH)3]+ (log b2,0,-3: -11.88).14 Since someinteraction between half-sandwich Ru2+ and Cl- can exist, allthe calculated log b values in this study are conditional stabilityconstants and are strictly valid for a medium of 0.20 M (KCl).(Detailed investigations on the half-sandwich Ru2+–Cl- system arein progress in our laboratory.)

For 1H-NMR titrations spectra were recorded on a BrukerAM360 FT-NMR instrument at 25 ◦C. Chemical shifts arereported in ppm (dH) from sodium 2,2-dimethyl-2-silapentane-5-sulfonate (Aldrich) as internal reference. Titrations were carriedout in D2O (99.8%) at cligand = 0.004 M using 1 : 1 or 1 : 4 M : L ratiosin order to register the pH* dependence of the chemical shifts ofthe CH, CH2 or CH3 nuclei of samples containing the various

ligands and [Ru(h6-p-cym)(H2O)3]2+. pH* was set up with NaODor DCl in D2O. Individual samples with different pH* values wereequilibrated for 60 min before measurements. pH* values (directpH-meter readings in a D2O solution of a pH-meter calibratedin H2O according to Irving et al.18 were converted to pH valuesmeasurable at an ionic strength of 0.20 M using the followingequation: pH = pH* + 0.40.

ESI-TOF MS analysis both in the positive and negative modewas carried out on a Bruker BIOTOF II ESI-TOF instrument.The measurements were performed on aqueous solutions at0.001 M [Ru(h6-p-cym)Cl2]2 concentration, at pH 3.5 and 10.0.The temperature of drying gas (N2) was 100 ◦C. The pressureof the nebulizating gas (N2) was 30 psi. Voltages applied at thecapillary entrance, capillary exit and the first and the secondskimmers were -4500, 120, 40 and 30 V, respectively. The spectrawere accumulated and recorded by a digitalizer at a sampling rateof 2 GHz. The spectrometer was operated at unit mass resolutionand was calibrated to sodium trifluoroacetate.

Results and discussion

Half sandwich Ru(II) complexes of (O,O) donor ligands as po-tential antitumor agents may not remain intact when entering thebody. Therefore, their features in aqueous medium and the strengthof the [Ru(h6-arene)]2+–ligand interaction are important to study.In order to estimate the stability constants of the complexes formedthe solution behavior of the metal ion, [Ru(h6-p-cym)(H2O)3]2+,hereafter abbreviated as Ru2+, in water was previously studied indetail and the composition and stability constants of the hydroxocomplexes were determined.14

Protonation constants of the ligands

The protonation constants of the (O,O) donor ligands (Chart 1)were determined from pH-potentiometric titrations in order tocheck their purity and to determine the exact concentration ofthe stock solutions that were used for the metal ion containingsystems. Evaluation of the titration curves yielded the log Kvalues appearing in Table 1. For Hsal and Htir the protonationprocesses belonging to the phenolate groups are beyond theaccurately measurable pH-range (log K ~ 13.4 for sal17 and 12.2for tir21) therefore these ligands are regarded as monoprotic acidsin this study. The re-determined protonation constants are in goodagreement with previous values obtained at identical experimentalconditions.20–24 For H2cbd the values also show good agreementwith reported data (log K1 = 5.46, log K2 = 2.85; T = 298 K, I =0.10 M).25

H+–[Ru(g6-p-cym)(H2O)3]2+–ligand systems

The studied (O,O) donor ligands (Chart 1) can be classified intothree groups on the basis of the solution behaviour against themetal ion therefore the results will be presented accordingly.

Hdhp, Hmalt, Hkoji, Hacac and Htir showed similar metal ionbinding capabilities and behaved in the same manner as it wasfound previously with the monohydroxamic acid, Hmeaha.14 Asa representative example the titration curves registered in the dhpsystem are presented in Fig. 1A. As it is seen the complex formationstarts well below pH 2.5 (like with Hmalt, Hkoji, Hacac but withHtir above 2.8) and regardless of the metal ion to ligand ratio only

10274 | Dalton Trans., 2010, 39, 10272–10278 This journal is © The Royal Society of Chemistry 2010

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online

Table 1 Stability constants of the proton (log K) and [Ru(h6-p-cym)]2+ (log b)a complexes of the ligands at 25.0 ◦C and I = 0.20 mol L-1 (KCl);3s standard deviations are in parentheses

dhp malt koji acac tirb cbd ox lact salb meahac

log KHA 9.78(1) 8.44(1) 7.69(1) 8.77(1) 7.48(1) 5.37(1) 3.75(1) 3.67(1) 2.69(1) 8.68log KH2A 3.68(1) — — — — 2.86(1) 1.3(2) — — —pHmax 11.0 8.5 4.0 4.0 11.0 6.4 3.5 5.5 6.3 11.0log bRuA 11.86(1) 9.05(1) 7.92(1) 8.56(1) — 4.05(1) 5.60(2) 2.79(1) 2.32(3) 9.79log bRuAH-1

1.83(1) -0.29(5) —d —d 2.67(1) -2.3(2) —d -2.25(1) -2.32(8) 0.42log bRuAH-2

— — — — -8.21(1) — — -7.66(1) -7.52(6) —pKRuA 10.03 9.34 —d —d — 6.35 —d 5.04e 4.64e 9.37pKRuAH-1

— — — — 10.88 — — 5.41 5.20 —log K*f 2.08 0.61 0.22 -0.21 -4.81 -4.18 0.53 -5.92 -5.01 1.11

a log b values refer to the equilibrium pRu + qA + rH �RupAqHr where Ru stands for [Ru(h6-p-cym)]2+. b As the phenolic groups deprotonate beyond theaccurately measurable pH-range (log K = 12.2 for tir21 and 13.4 for sal20) these ligands are regarded as monoprotic acids in this work. c Ref. 14. d Accuratevalue could not be determined due to slow equilibria above pH ~ 4. e Refers to the deprotonation of the alcoholic (lact) or phenolic (sal) OH group ratherthan that of the co-ordinated water. f Refers to equilibria 1–3 (see text).

Fig. 1 Titration curves with (A) dhp, (B) cbd and (C) lact, for (a) theH+–ligand systems and the [Ru(h6-p-cym)(H2O)3]2+–ligand systems at (b)1 : 2 and (c) 1 : 1 ratios.

1 : 1 complexes are formed. This is supported by the fact that bypH ~ 6 one (for Hmalt and Hkoji slightly less than one) equivalentof extra base is consumed at 1 : 1 ratio due to the metal ion assisted

deprotonation and coordination of the ligands. In all the systemsafter the jump in pH (for Hdhp above pH 9.2, Hmalt 8.4, Hkoji8.2, Htir 10.0) a new base consumption process is detectable thatcan be assigned to the deprotonation of the water molecule in thecoordination sphere of the metal ion. Evaluation of the titrationcurves for these systems using different metal ion to ligand ratiosresulted in the models which are summarized in Table 1 togetherwith the determined stability constants of the complexes. Inagreement with previous observations for meaha,14 the formationof [RuA] and [RuAH-1] can be detected with these ligands. ForHacac and Hkoji, however, slow processes were detected abovepH ~ 4.0, therefore reliable stability constant for [RuAH-1] couldnot be calculated. ESI-TOF-MS studies carried out at differentpH values and metal ion to ligand ratios (Table 2) revealed thealmost exclusive formation of mononuclear complexes in aqueoussolution, and therefore it is plausible to assume that in [RuA] theligands coordinate via an (O,O) chelate (structure II in Scheme 1),and in [RuAH-1] the deprotonation of the water molecule in thecoordination sphere occurs (structure III). As Htir is regarded asa monoprotic acid (vide supra) the stoichiometry [RuAH-1] mayrefer in this system to the (O,O) chelated species (structure II) and[RuAH-2] to the mixed hydroxo complex (structure III). As anexample, the calculated concentration distribution curves for theRu2+–dhp system appear in Fig. 2A.

Scheme 1

This journal is © The Royal Society of Chemistry 2010 Dalton Trans., 2010, 39, 10272–10278 | 10275

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online

Table 2 Experimental m/z values corresponding to the various uncomplexed, complexed and hydroxo species identified at MS conditions in the acidicand basic pH-range

dhp malt koji acac lact sal cbd ox

Acidic [Ru(cym)Cl]+ — — — — 270.981 270.980 270.981 270.980[Ru(cym)Cl(H2O)]+ — — — — 288.992 288.993 288.998 288.982[{Ru(cym)Cl}2(m2-Cl)]+ — — — — 576.933 576.933 — —[Ru(cym)A]+ 374.070 361.040 377.036 335.059 — — — —H[Ru(cym)A]+ — — — — — — 379.045 325.001K[Ru(cym)A]+ — — — — — — 417.007 362.955H[Ru(cym)AH-1]+ — — — — 325.036 373.038 — —

Basic [Ru(cym)A]+ 374.074 361.040 377.033 335.057 — — — —[{Ru(cym)A}2(m2-OH)]+ 765.155 739.089 — — — — — —K[Ru(cym)A(OH)]+ 430.041 417.008 433.000 391.025 — — — —[{Ru(cym)}2(m2-OH)3]+ — 523.042 523.038 523.040 523.038 523.043 523.037 523.037

Fig. 2 Calculated concentration distribution curves for the[Ru(h6-p-cym)(H2O)3]2+–dhp (A), –cbd (B) and –lact (C) systems at1 : 1 ratio. cRu = 0.002 M.

With the two dicarboxylic acids, H2cbd and H2ox, a slightlydifferent complexation scheme was found. As an example, thetitration curves registered in the cbd system are shown in Fig. 1B.It can be seen that the complex formation starts at pH ~ 3.1 (H2ox:2.3). Unlike the previous ligands, here – most probably due to thelow basicity of the donor groups of the ligands – complexationand hydrolysis of the metal ion take place in overlapping processesbelow pH ~ 7.0 and no further base consumption is detectable afterthe jump in pH above ~ 10. Evaluation of the potentiometric dataobtained below pH 6.5 (H2ox: 3.5) resulted in the compositionsand stability constants summarised in Table 1, while the species

distribution curves for the Ru2+–cbd system at 1 : 1 ratio are shownin Fig. 2B. Above pH 6.5 (H2ox: 3.5) slow processes were detectedin both systems attributable to the hydrolysis of the (O,O) chelatedspecies yielding the dinuclear hydroxo complexes of Ru.

Hlact and Hsal were found to bind the metal ion too, but asthe titration curves for the Ru2+–lact system reveal (Fig. 1C),1.5 equivalents of extra base is consumed at 1 : 1 ratio withthese ligands. This unambiguously supports the metal ion assisteddeprotonation and coordination of the alcoholic OH (for Hsalphenolic OH) group to yield [RuAH-1] (structure II). Furthermore,formation of either a dinuclear mixed hydroxo complex with[Ru2A2H-3]- stoichiometry or hydrolysis resulting in [Ru2(OH)3]+

(structure IV) can occur (as a further 0.5 equivalent of extrabase is consumed). ESI-MS measurements, however, showed noindication of the formation of the [Ru2A2H-3]- complexes thereforethe formation of IV is more likely. Since above pH 6.0 slowprocesses were detected and no pH equilibrium could be reachedwithin 15 min, only data below pH ~ 6.0 were used for thecalculations. Treatment of the data with PSEQUAD resulted inthe models and stability constants shown in Table 1. [RuA]+ mayrefer to a complex in which the ligand is deprotonated at thecarboxylic function, while in [RuAH-1] an (O,O) chelated ligandand one water molecule take the coordination sites (structure II).[RuAH-2]- (more precisely [RuAH-1(OH)]-) is most probably amixed hydroxo complex (structure III). Both the stability constantsfor [RuA]+ and the values of pKRuA suggest that in [RuA]+

beside the COO- function the protonated alcoholic/phenolic OHgroup also coordinates yielding a chelate. Identical coordinationmode was found previously in the VO(IV)-lactate system.24 As anexample, the species distribution curves for the Ru2+–lact systemare shown in Fig. 2C.

1H-NMR studies for these systems at 1 : 1 and 1 : 4 ratios pro-vided further proof for the models outlined above. pH dependenceof the chemical shift values of doublets belonging to the p-cymene–CH(CH3)2 protons turned to be an excellent tool to follow thecomplexation in the systems studied. As a representative examplethe doublets detected in the [Ru(h6-p-cym)(H2O)3]2+–malt systemat 1 : 1 metal ion to ligand ratio are shown in Fig. 3, while thefull spectra can be seen in Fig. S1, ESI.† At pH ~ 2.38 besidesthe uncomplexed [Ru(h6-p-cym)(H2O)3]2+ at 1.32 ppm14 a newdoublet at 1.26 ppm is attributable to [Ru(h6-p-cym)(malt)(H2O)]+.On increasing the pH the first signal gradually decreases anddisappears by pH ~ 4.9 at 1 : 1 ratio. Parallel with this the intensityof the new signal at 1.26 ppm increases first and remains steady

10276 | Dalton Trans., 2010, 39, 10272–10278 This journal is © The Royal Society of Chemistry 2010

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online

Fig. 3 Dependence on pH of the high field region of the 1H-NMRspectrum of a solution containing [Ru(h6-p-cym)Cl2]2 and Hmalt in a1 : 1 ratio in D2O at 298 K. cdimer = 1.00 mM.

above pH 5.0. At pH ~ 10.3 an upfield shift of this signal canbe seen, which strongly suggests some minor structural change,e.g. deprotonation. Similar behaviour was found in the [Ru(h6-p-cym)(H2O)3]2+–meaha system.14 Besides the signal at 1.26 ppm, anew one appears above pH 8.0 at 1 : 1 ratio, which was assignedto [{Ru(h6-p-cym)}2(m2-OH)3]+ according to the value of thechemical shift (1.22 ppm).14 Similar observations can be made withall the ligands regarding the pH dependence of the NMR signalsat 1 : 1 or 1 : 4 ratios; the results for 1 : 1 ratios are summarized inFig. 4.

Fig. 4 Tentative formation pH-ranges for [Ru(h6-p-cym)(H2O)3]2+ ( ),(O,O) chelated complexes ( ) and [{Ru(h6-p-cym)}2(m2-OH)3]+ ( ) ac-cording to 1H-NMR data.

ESI-TOF-MS in the positive mode was also used to assist inidentifying the pH dependence of the speciations for the various

[Ru(h6-p-cym)(H2O)3]2+–(O,O) donor systems. The major experi-mental m/z values corresponding to [Ru(h6-p-cym)(H2O)3]2+, thecomplexed species, [{Ru(h6-p-cym)}2(m2-OH)3]+ or other hydroxospecies present at MS conditions are summarized in Table 2. Allof the registered spectra exhibit the correct isotopic patterns (as itwas proved with the comparison of the simulated spectra). Somerepresentative examples can be found in Fig. S2, ESI.†

Conclusion

Similarly to the [Ru(h6-p-cym)(H2O)3]2+–monohydroxamate–(O,O) interaction, recently found by us,14 the (O,O) donor ligandsused in this study having different basicity of the chelating oxygenatoms react in fast processes with the metal ion at acidic conditions.The starting pH and the extent of the complex formation inaqueous solution depend upon the basicity of the donor atomsand the ring size of the chelate formed.

Since the basicity of the studied ligands varies considerably(Table 1), and the composition of the (O,O) chelated complexeshaving water molecules in the remaining third position of the metalion (structure II) can also be different ([RuA] vs. [RuAH-1] forlact, sal and tir) the direct comparison of the calculated stabilityconstants is not possible. In order to obtain direct informationfor the stability of the (O,O) chelates formed with the differentligands the constants (log K*) referring to the following protondisplacement equilibria were calculated:

Ru2+ + HA � [RuA]+ + H+ (dhp, koji, malt, acac) (1)

Ru2+ + H2A � [RuA] + 2H+ (ox, cbd) (2)

Ru2+ + HA � [RuAH-1] + 2H+ (lact, sal, tir) (3)

where Ru2+stands for [Ru(h6-p-cym)(H2O)3]2+. The larger the logK* value, the more stable is the (O,O) chelate formed with themetal ion.

The obtained log K* values are summarized in Table 1. Fromthe comparison of the derived equilibrium constants the followingconclusions can be drawn: (i) among the chelates with similar type(e.g. carboxylate) donor atoms the five-membered ones are morestable than the six-membered ones (ox. vs. cbd). (ii) Delocalisationof the negative charge of the (O,O) chelate results in slightly“pseudo aromatic” character leading to extra stability for themetal complex (e.g. acac, dhp). This is clearly seen when thederived constant obtained for dhp with those for malt or koji arecompared (Table 1): for the chelate with dhp at least 1.5 log unitextra stability can be found. (iii) Monohydroxamates capable offorming five-membered (O,O) chelates (log K* = 1.11 for meaha)14

are almost as strong chelators for [Ru(h6-p-cym)(H2O)3]2+ as dhp.(iv) Ligands with low basicity O donor atoms (ox, cbd) bind[Ru(h6-p-cym)(H2O)3]2+ at acidic conditions but at elevated pHvalues are not able to prevent the hydrolysis of the metal ionand the formation of the more inert [{Ru(h6-p-cym)}2(m2-OH)3]+

occurs at physiological pH in these systems. On the contrary, withtir the interaction starts only at weakly acidic pH but the chelatesproved to be significantly more resistant against hydrolysis at basicconditions.

This journal is © The Royal Society of Chemistry 2010 Dalton Trans., 2010, 39, 10272–10278 | 10277

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online

Acknowledgements

We thank members of the EU COST Action D39 for motivatingdiscussions. This work was supported by the Hungarian ScientificResearch Fund (OTKA K76142).

References

1 P. C. A. Bruijnincx and P. J. Sadler, Adv. Inorg. Chem., 2009, 61, 1–62.

2 R. Fernandez, M. Melchart, A. Habtemariam, S. Parsons and P. J.Sadler, Chem.–Eur. J., 2004, 10, 5173–5179.

3 M. Melchart, A. Habtemariam, S. Parsons, S. A. Moggach and P. J.Sadler, Inorg. Chim. Acta, 2006, 359, 3020–3028.

4 A. F. A. Peacock, M. Melchart, R. J. Deeth, A. Habtemariam, S.Parsons and P. J. Sadler, Chem.–Eur. J., 2007, 13, 2601–2613.

5 W. Kandioller, C. G. Hartinger, A. A. Nazarov, M. L. Kuznetsov,R. O. John, C. Bartel, M. A. Jakupec, V. B. Arion and B. K. Keppler,Organometallics, 2009, 28, 4249–4251.

6 W. H. Ang, E. Daldini, C. Scolaro, R. Scopelliti, L. Juillerat-Jeanneratand P. J. Dyson, Inorg. Chem., 2006, 45, 9006–9013.

7 C. Gossens, A. Dorcier, P. J. Dyson and U. Rothlisberger,Organometallics, 2007, 26, 3969–3975.

8 W. H. And, Z. Grote, R. Scopelliti, L. Juillerat-Jeannerat, K. Severinand P. J. Dyson, J. Organomet. Chem., 2009, 694, 968–972.

9 A. Levina, A. Mitra and P. A. Lay, Metallomics, 2009, 1, 458–470.

10 Y. K. Yan, M. Melchart, A. Habtemariam and P. J. Sadler, Chem.Commun., 2005, 4764–4776.

11 C. G. Hartinger and P. J. Dyson, Chem. Soc. Rev., 2009, 38, 391–401.12 A. Bergamo, A. Masi, A. F. A. Peacock, A. Habtemariam, P. J. Sadler

and G. Sava, J. Inorg. Biochem., 2010, 104, 79–86.13 M. Liu, Z. J. Lim, Y. Y. Gwee, A. Levina and P. A. Lay, Angew. Chem.

Int. Ed., 2010, 49, 1661–1664.14 P. Buglyo and E. Farkas, Dalton Trans., 2009, 8063–8070.15 M. A. Bennett and A. K. Smith, J. Chem. Soc., Dalton Trans., 1974,

233–241.16 A. Schluter, K. Bieber and W. S. Sheldrick, Inorg. Chim. Acta, 2002,

340, 35.17 G. Gran, Acta Chem. Scand., 1950, 4, 559.18 H. M. Irving, M. G. Miles and L. D. Pettit, Anal. Chim. Acta, 1967,

38, 475.19 L. Zekany, I. Nagypal, in: D. Leggett, (Ed.), Computational Methods

for the Determination of Stability Constants, Plenum, New York, 1985.20 M. Jezowska-Bojczuk, H. Kozlowski, A. Zubor, T. Kiss, M. Branca, G.

Micera and A. Dessı, J. Chem. Soc., Dalton Trans., 1990, 2903–2907.21 T. Kiss, I. Sovago and R. B. Martin, J. Am. Chem. Soc., 1989, 111,

3611.22 P. Buglyo, T. Kiss, E. Kiss, D. Sanna, E. Garribba and G. Micera,

J. Chem. Soc., Dalton Trans., 2002, 2275–2282.23 P. Buglyo, E. Kiss, I. Fabian, T. Kiss, D. Sanna, E. Garribba and G.

Micera, Inorg. Chim. Acta, 2000, 306, 174–183.24 G. Micera, D. Sanna, A. Dessı, T. Kiss and P. Buglyo, Gazz. Chim. Ital,

1993, 123, 573–577.25 T. Rau, M. Shoukry and R. van Eldik, Inorg. Chem., 1997, 36, 1454.

10278 | Dalton Trans., 2010, 39, 10272–10278 This journal is © The Royal Society of Chemistry 2010

Dow

nloa

ded

by U

nive

rsity

of

Vir

gini

a on

17

Sept

embe

r 20

12Pu

blis

hed

on 0

4 O

ctob

er 2

010

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C0D

T00

469C

View Online