J . CHEM. SOC. DALTON TRANS. 1982 865Structural and Mechanistic Studies of Co-ordination Compounds. Part33.’ Inner-sphere vs. Outer-sphere Mechanisms in the Reductions ofSome trans-Dianiono(tetramine)ruthenium( 111) Cations by Chromium(ii)and Vanadium(i1)By Chung-Kwong Poon,’ Tin-Wu Tang, and Tai-Chu Lau, Department of Chemistry, University of HongKong, Pokfulam Road, Hong KongThe kinetics of the reduction of trans-[RuLlAX]+ [L’ = bis(ethane-l,2-diamine); AX = C12, Br2, la, or BrCI] andtrans- [ RuLCI,] + [ L = L2 (3,7-diazanonane-l,9-diamine), L3 (1,4,8,11 -tetra-azacyclotetradecane), L4 (1,4,8,12-tetra-azacyclopentadecane), L5 or L6 (C-meso or C-rac-5,5,7,12,12,14-hexamethyl-l,4,8,1 I -tetra-azacyclotetra-decane respectively)] by chromium(l1) and vanadium(l1) have been studied at 25.0 “C in aqueous acidic solutionsof ionic strength 0.50 mol dm-3.The chromium(l1) reductions have been shown to proceed by an inner-spheremechanism via the bridging ligand X-. These reactions are very sensitive to steric effects with trans-[RuL5CI2]+and trans-[RuLW,]+ more reactive than the corresponding steric-free L3 complex by a factor of 76.5 and 58.0respectively. For trans- [Ru L1 AX] + complexes, the second-order rate constants drop systematically in the followingorder of AX : l2 > Br2 > ClBr > BrCl > CIS (148, 38, 37, 31, and 15 dm3 mol-1 s-1 respectively, after a statisticalfactor of two has been applied to the dihalogeno-complexes). For vanadium(l1) reductions, the second-orderrate constants, ranging from 1.21 x 103 dm3 mol-l s-l for trans-[RuL1C12]+ to 1.27 x l o 4 dm3 mol-1 s-1 fortrans- [RuL’ I,]+, are much greater than the ligation rates of vanadium(l1). This, together with the observationthat steric factors have relatively little effect on the reactivity, strongly supports an outer-sphere mechanism forthese vanadium(l1) reductions.THE study of reductions of ruthenium(rI1) amine com-plexes has so far been confined to ammine systems, suchas [Ru(NH,),],+,~ [Ru(NH3),X]nf (X = halide, H20, orcarboxylate) ,3 and cis- and trans-[Ru(NH,),ClX]n+(X = C1 or H,0).4 We have recently reported5-7 thesyntheses of extensive series of amine complexes of thetype trans-[RuLAX]+, where L represents either twobidentate or one quadridentate amine and A and X areunidentate ligands, and have also reported the chelationeffects of L on chromium(1r) reduction of trans-[RuLCl,]+[L = L1, bis(ethane-1,Z-diamine) ; L2, 3,7-diazanonane-1 ,g-diamine, or L3, 1,4,8,1l-tetra-azacyclotetradecane].L’vL2 L3These reactions have been shown to proceed by an inner-sphere mechanism.8 In this paper, we have extendedthe study to a much wider range of complexes, trans-[RuLIAX]+ (AX = Br,, I,, or ClBr) and trans-[RuLCl,]+(L = L4, 1,4,8,12-tetra-azacyclopentadecane, L5 or L6,C-meso or C-rac-5,5,7,12,12,14-hexamet hyl- 1,4,8,11-tetra-azacyclotetradecane respectively), in order tounderstand steric and ring-size effects and the effects ofco-ordinated halides on these reduction reactions.Reductions by vanadium(@ give rise to both inner-and outer-sphere reaction^.^^^^^^^ I t was, therefore, alsothe intention of this work to investigate the vanadium(11;reduction mechanisms of the above series of trans,tetramineruthenium(Ir1) and trans-[RuLCl,]+ (L = L1L2, or L3) complexes.EXPERIMENTALThe complexes trans-[RuLIAX][C1O,] (AX = Cl,, Br,,or BrCl), trans-[R~L11,]1,~ tr~ns-[RuL~Cl,][ClO~,~~ trans-[ R U L ~ C ~ , ] C ~ , ~ ~ ~ and truns-[RuLClJ[ClO J 7 (L = L4, L6,or L6) were prepared according to published methods.Chromium(I1) solutions were prepared by reducing potas-sium dichromate solutions first to chromium(II1) withhydrogen peroxide and then to chromium(I1) with amal-gamated zinc.12a Chromium(I1) was analysed by oxidizingwith deoxygenated iron(rI1) and determining the reducediron(I1) with acid dichromate.Vanadium(I1) solutions wereprepared by reduction of solutions of vanadium(v) oxide intoluene-p-sulphonic acid with amaJgamated zinc under anatmosphere of argon and were used immediately afterstandardization with iron(II1) in the presence of excess ofthiocyanate.1s The ionic strength was maintained withsodium toluene-p-sulphonate.Kinetic Measurements.-All kinetics were followedspectr ophotometrically in situ using an Aminco-Morrowstopped-flow spectrophotometer equipped with an AmincoDASAR (data acquisition, storage, and retrieval) system.Experimental details on data collection, temperature con-trol, and data treatment have been described previously.l3All operations were carried out under deoxygenated argon.Syringe techniques were used for the transfer of air-sensitivesolutions866 J.CHEM. SOC. DALTON TRANS. 1982RESULTSThe spectrophotometric changes for both chrumiuni(i1)and vanadium(I1) Ieductions in toluene-p-sulphonic acid(Hpts, 0.1 mol dm-3) are characterized by a gradual dis-appearance of the intense ligand-to-metal charge-transferbands of the complexes, thus confirming the conversion ofd5 ruthenium(II1) centre into the d6 ruthenium(I1) counter-part. For inner-sphere chromium(I1) reductions, * thenature of the bridging ligand in the reaction of trans-[RuLICIBr]+ is not trivial. I t was determined by thefollowing method. An acidic solution (0.1 mol dm-3Hpts) of trans-[RuLICIBr]+ at 25.0 "C was treated with asuitable quantity of chromium(I1) for 1.5 min.Air wasimmediately admitted to re-oxidize the ruthenium(I1)instantaneously back to the inert ruthenium(II1) species.The U.V. spectrum of this solution, labelled A, showed that i twas a mixture of trans-[RuLICI (OH,)I2+ and trans-[RuLIBr-(OH2)I2+ indicating that both C1- and Br- are functioning asbridging ligands in a competitive manner. To determinethis competition ratio quantitatively, solution A wasequally divided into two portions, A1 and A2. Portion A1was treated with an excess of NaCl and the solution warmedto ca. 40 "C for ca. 20 min. This gave a mixture of trans-[RuLICIBr]+ and trans-[RuL1C12]+, the exact concentrationsof which were determined from the known molar absorptioncoefficients of these species a t A,.(343 nm) of the latter.In a typical run, the concentrations of the chlorobromo- anddichloro-complexes were found to be CClBr = 8.2 >( andccI, = 9.6 x low5 mol dm-3 respectively. Similarly, solu-tion A2 was treated with excess of NaBr and the final solu-tion, as determined from the absorbances a t 409 nm,corresponded to a mixture of trans-[RuLICIBf]+ (cCIBr =9.8 x mol dm-3) and trans-[RuLIBr,]+ (cBr, = 8.0 x10-5 mol dm-3).Assuming the following competitive reductions of trans-[RuLICIBr]+ by chromium(I1) [equations (1) and (2)], whereThe kinetics of the redox reactions at 25.0 "C were followedspectrophotometrically a t the absorption maxima of theruthenium(II1) complexes. Concentrations of the reduct-ants, either chromium(I1) or vanadium(II), were kept atleast 20-fold (1 .O x 10-3-1 .O x lo-, mol dm-3) greater thanthose of the oxidants (0.80 x 10-4-4.3 x rnol dm-3).Second-order rate constants a t 25.0 "C for chromium(I1) andvanadium(I1) reductions of some trans-[RuLAX]+complexes in 0.10 mol dm-3 toluene-p-sulphonic acidand a t I = 0.50 mol dm-3 (sodium toluene-p-sulphon-ate) a1 O-lkc, 1 0-3kv Complexr-A____.L A x' dm3 mol"' s-1 dm3 mol-1 s-1(NHJ4 C1 C1L1 c1 c1L' Br C1L' C1 BrL' Br BrL' I IL2 d c1 c1L3 c1 c1L4 c1 c1L5 c1 c1L6 c1 c112.6 b3.023.13.77.6329.54.736.46 c11.64943750.83 b1.212.252.253.072.033.483.567.287.6912.7a Second-order rate constants, being independent of acid(0.05-0.50 mol dm-3), are obtained from the slopes of theleast-squares plots of kobs.us. [reductant] over the range1.0 x 10-3-1.0 x rnol dm-3. In the case of chrom-ium(I1) reactions, X- represents the bridging ligand. Ref. 4,I = 0.10 mol dm-3. C Ref. 8. dR,S-isomer. e These twocomplexes are unstable in very dilute acids. Rate constantsare independent of acid in the range 0.20-0.45 mol dm-3.The ionic strength of the reaction solutions was maintaineda t 0.50 mol dm-3. Semi-logarithmic plots of lo@, - A , )'us. time, where At and A , represent absorbances a t time Iand a t infinity (10 half-lives) respectively, were linear over atleast three half-lives. The pseudo-first-order rate con-stants, Kobs., are independent of the wavelengths of measure-kClBr - - trans-[RuL1C1(OH2)]+ + [Cr(OH,),Br]2+ (1)(2)kBrC1 trans-[RuLICIBr]+ + [Cr(OH,),I2+ - 1- trans- [RuLlBr (OH,)]+ + [Cr (OH,) 5C1] 2+kAX represents the appropriate second-order rate constant forthe bridging pathway (X = C1 or Br), i t is obvious that thecompetition ratio of k*Rr/kBrCl is given by the concentrationratio of [RuL1Cl(OH,)+]/[RuL1Br(OH2)+]. In solution Al,CC~, and GCIB~, in fact, corresponded to the concentrations oftrans-[ RuLIC1( OH ,)I + and trans- [RuLlBr (OH,)]+ respect-ively of the mother solution A, and hence give the competi-tion ratio k c 1 B r / k B ~ = 1.17.This agrees very well withthe independently determined competition ratio of 1.23 fromsolution A2 where CCIB~ and cBr, denote the concentrations oftrans-[RuLIC1(OH,)]+ and trans-[RuLIBr(OH,)]+ respect-ively. The consistency of these two ratios further indicatesthat the extent of aquation of the reduced species, trans-[RuLIA(OH,)]+ (A = C1 or Br), is insignificant within thetime scale of the experiment.However, when the experi-ments wererepeated with an extended reduction time, it wasfound that this competition ratio for solution A1 graduallyincreased but that for solution A2 decreased with the reduc-tion time. This observation is consistent with the aquationof the reduced species thus generating some trans-[RuL1-(OH,),I2+ in solution A which eventually gives a higherestimate of dihalogeno-complexes over trans-[RuLICIBr]on reaction with excess of halide.ment, acid (0.05-0.50 mol dm-3 for L1-4 complexes and0.20-0.45 mol dm-3 for L5 and L6 complexes), and sub-strate concentrations, b u t increase linearly with reductantconcentrations. For each substrate, experiments wererepeated for a t least five different concentrations of thereductants.Second-order rate constants, kcr or k v , wereobtained from the slopes of the appropriate linear plots ofRobs. vs. [reductant] by the method of least squares.The rate constants for the bromide-bridged and chloride-bridged paths, kclRr and K B ~ c ~ respectively, of trans-[RuLICIBr]+ were determined from the competition ratio(average k ~ ~ B ~ / k ~ & l = 1.20) and equation (3), where K , is= KBrCl + KClBr (3)the observed second-order rate constant for the disappear-ance of Irans-[RuLICIBr]+ in the chromium(I1) reductionreactions.All these kinetic data are collected in the Table.DISCUSSIONIt has been shown unambiguously that chromium(I1)reductions of tram-[RuLCl,]+ (L = L1-3) proceed by aJ. CHEM.SOC. DALTON TRANS. 1982 867inner-sphere mechanism according to the generalequations (4) and ( 5 ) , with the second-order rate con-stants kcr given by equation (6). The present observ-ation of steric acceleration (trans-[RuL5C12]+ and[ALRuX]+ -/- [Cr(OH2)6]2f +[ALRu-X-Cr( OH,)5]3+K[ALRu--X-Cr(OH,),l3+ (4)k[RULA(OH,) I + + [Cr (OH,) 5x1 2+ (5)kcr = k K (6)trans-[RuL6C1,] + are more reactive than trans-[RuL3C12] +by a factor of 76.5 and 58.0 respectively) is fully con-sistent with this mechanism.I t is well known thatsteric effects accelerate the dissociation of a leavinggroup. The acid hydrolysis of trans-[CoLCl,]+ (L =L5v6) has been shown to be faster than that of trans-[CoL3C12]+ by a factor of ca. 103.14915 This steric effectis even more pronounced for the correspondingruthenium(II1) system.16 It seems reasonable to expectthat the values of k in equation (5) for the L5 and Lschloride-bridged intermediates are greater than that ofthe corresponding L3 intermediate by at least a factor ofca. lo3. The presence of [Cr(OH,),C1I2+ as a leavinggroup, which is larger than C1-, would only serve toenhance the steric acceleration on k . The variation ofK in equation (4) with additional methyl groups in themacrocycle L is probably associated with solvation andnon-bonding steric crowding effects.The association oftwo charged species into a charged dimer is probablyaccompanied by a decrease in solvation energy. Thelarger starting complex trans-[RuLCl,] + (L = L59 6,would suffer a smaller relative loss in solvation energy bydesolvation than trans-[RuL3Cl,] +, and hence wouldassume a greater value of K . On the other hand, theincreased non-bonding steric crowding of trans-[RuLCl,] +with additional methyl groups would reduce the value ofK . The interplay of these two opposing effects, with thelatter probably more important for L5 and L6 systems,might lead to an overall decrease in the value of K ,although to a much less extent as k increases with stericeffects. The combination of k and K , i.e. kc,, is there-fore expected to increase with steric effects.Theobserved steric ratios of 76.5 and 58.0 are consistent withthis inner-sphere mechanism.As far as ring-size effect is concerned, the observationthat trans-[RuL*Cl,]+ is slightly more reactive thantrans-[RuL3C1,]+, by a factor of 1.8, is also consistentwith the inner-sphere mechanism. This effect, whichhas the same magnitude as the chelation effect (trans-[RuL3Cl,]+ is more reactive than trans-[RuL2C1,]+ by afactor of 1.4), probably arises from the same solvationeffects and steric constraints.*A comparison of kcr for various trans-[RuLIAX]+ com-plexes, after correcting for the statistical factor of two fordihalogeno-complexes, shows that the second-order rate* An upper limit of ca. 40 dm3 rno1-I s-1 for vanadium(r1)inner-sphere reactions has been estimated by Sutin."constants decrease systematically with the nature ofAX : I, > Br, 2 ClBr >, BrCl > C1, (148, 38, 37, 31,and 15 dm3 mol-l s-l respectively). It is worth noting,even based on the limited number of experimental data,that the bridging efficiencies of the halides depend on thenature of the trans-activating ligands.Thus, withchloride as the trans-activating ligand, bromide functionsas a better bridge than chloride (kmr/kcl, = 2.5),whereas for trans-bromo-complexes the bridging efficien-cies of bromide and chloride are virtually identical(kBr,/kBa = 1.2). On the other hand, the trans-activating efficiencies of the halides also depend on thenature of the bridging ligands.With chloride as thebridging ligand, bromide has a greater trans-activat ingeffect than chloride (kBfll/ka, = 2.1) but this superioritydisappears when bromide becomes the bridging ligand(kBr,/kClBr = 1.0). It thus appears that the order ofbridging or trans-activating efficiencies of halides for oneruthenium(Ir1) system may not be the same for another,even though very similar, and care must be exercised indiscussing these efficiencies. The much greater reac-tivity of trans-[RuL112]+ relative to other L1 halogeno-complexes reported here could then be due to either agreater trans-activating effect or bridging effect of iodideor both. The present experimental results do not allowus to draw a conclusion on the relative merits of theseeffects.The behaviour of vanadium( 11) reductions of thesetetramineruthenium(II1) complexes appears to be quitedifferent to that of the chromium(I1) reductions dis-cussed above.The fact that all the second-order rateconstants, k v , are much greater than the ligation ratesof vanadium(I1) * clearly indicate an outer-spheremechanism for these reactions. The lack of ring-sizeand significant steric effects in contrast to those observedfor inner-sphere chromium(I1) reactions further supportsthis mechanism. That the bulkier trans-[RuLCl,] +(L = L59'7 complexes are slightly more reactive thantrans-[RuL3C1,]+ by a factor of ca. 2.1 is probably asolvation effect.We thank the Committee on Research and ConferenceGrants of the University of Hong Kong for support.[1/1523 Received, 30th September, 19811REFERENCESDalton Trans., 1982, 531.1 Part 32, C. K. Poon, T. C. Lau, and C. M . Che, J . Chew&. SOC.,2 C. A. Jacks and L. E . Bennett, Inorg. Chem., 1974,18, 2035.3 J . A. Stritar and H. Taube. Inorg. Chem., 1969, 8, 2281.4 W. G. Movius and R. G. Linck, J . A m . Chem. Soc., 1970, 92,5 P. K. Chan, D. A. Isabirye, and C. K. Poon, Inorg. Chem.,6 C. K. Poon and C. M. Che, J . Chem. SOC., Dalton Trans.,7 C. K. Poon and C. M. Che, Inorg. Chem., 1981, 20, 1640. * C. K. Poon, T. W. Tang, and T. C. Lau, J . Chem. SOC.,s P . R. Guenther and R. G. Linck, J . Am. Chem. SOC., 1969,2677.1975, 14, 2579.1980, 756.Dalton Trans., 1981, 2556.91, 3769868 J. CHEM. soc. DALTON TRANS. 1982lo B. Grossman and A. Haim, J . Am. Chem. Soc., 1971, 98,6490. 1978, 216.C. K. Poon and D. A. Isabirye, J . Chem. SOC., Dalton Trans.,1977, 2115.Analysis,' 3rd edn., Longmans, London, 1961, (a) p. 340; ( b ) p.337. l7 N. Sutin, Acc. Chem. Res., 1968, 1, 225.13 C. K. Poon and P. W. Mak, J . Chem. Soc., Dalton Trans.,W. K. Chau and C. K. Poon, J . Chem. SOC. A , 1971, 3087.l 6 W. K. Chau, W. K. Lee, and C. K. Poon, J . Chem. SOC.,T. C. Lau, unpublished work.l2 A. I. Vogel, ' A Text-book of Quantitative Inorganic DaEton Trans., 1974, 2419