摘要:

1974Kinetics of Ternary Complex Formation between Nickel Species and5-N itrosalicylic AcidBy David N. Hague and Keith Kinley, University Chemical Laboratory, CanterburyStopped-flow and temperature-jump relaxation methods have been used to measure rate constants and activationparameters for the formation and dissociation of a 1 : 1 complex between nickel(r1) and 5-nitrosalicylate(2-)(nsa) and of ternary complexes between nsa and nickel(r1)-polytriphosphate, 4minodiacetate. -nitrilotriacetate,-et h y I e lied i a m i n e -NN’- d i a ceta te, -diet h y I en etr i a m i n e, -trie t h y I en ete tra m i n e, and -2,2’, 2”- tri a m i n o tri et h y I a minecomplexes. Rate constants and activation parameters have also been measured for the reaction between the NiIIcomplex of ethylenediamine-NN’-diacetate and pyridine-2-azo-p-dimethylaniline. The results can be rationalizedin terms of the normal dissociative model provided the local charge density on the metal species is consideredrather than its overall charge.The results provide further evidence that the number of binding nitrogen atoms onthe first ligand has a larger influence on ternary complex formation than its charge.THE kinetics of formation and dissociation of 1 : 1 com-plexes of nickel(I1) in water have been widely studiedsince the introduction of rapid-reaction techniques.Rate constants for the formation of such complexes, kf,are usually in the range 103-105 1 mol-l s-l and it isgenerally accepted that the mechanism of complex forni-ation with a unidentate ligand involves rapid diffusingtogether of the two species to form an outer-sphere com-plex, followed by rate-limiting replacement of a watermolecule in the inner co-ordination sphere of the metalion by the ligand. Where the incoming ligand containsa.second or further binding group, the rate-determiningstep is usually formation of the first metal-ligand bond.If KO, is the formation constant of the outer-sphere inter-mediate, it can be shown that, provided the concentra-tion of the intermediate is small, k f is given by KoSKex,where k,, is the (first-order) rate constant for waterexchange. In most cases the value of KO, obtained fromkf/k,, is comparable to that calculated on the basis of theFuoss equation,2 but low values have been rationalizedin terms of rate-limiting ring closure3 and high valuesin terms of an internal conjugate-base mechani~m.~Comparatively little is known about the influenceexerted by a ligand which is already bound to the nickel-(11) ion on the subsequent reaction of the latter withanother ligand, although it has been found5 that thereis a general increase in k,, as the number of bound ali-phatic nitrogen atoms increases.This labilisation of theremaining water niolecules by N atoms is reflected inincreased values of kf for the neutral incoming ligaiidsNH, 6 and pyridine-2-azo-$1-dimethylanilhe (pada) ,‘ andin both cases the presence of negatively charged oxygenatoms in the inner co-ordination sphere of nickel(I1)has little influence on hf.The purpose of the presentstudy is to investigate the effect of charge on the in-coming ligand. Rate constants and activation para-meters, as nieasured by the stopped-flow and tempera-1 See, for example, XI. Eigen and R. G. Wiikins, ‘ Mechanisfiiso f Inorganic Reactions,’ ed. R. 1:. Gould, Adu. Chew. Series, No.49, Amer. Chcm. Soc., Washington, D.C., 1965, p. 55; D. J.Hewkin and R. H. Prince, Co-ovdination Ckern. Rev., 1970, 5, 4.5;It. G. Wilkins, Accoitvlts Chem. Res., 1970, 3, 408.K. M. Fuosb, J . Amev. Chew. SOC., 1958, 80, 5059; see also1cf. 1 .3 See, for example, X. ICowalak, K. Kustin, R. F. Pasternack,and S. Petrucci, j . Anirr.. ChewL. SOC., 1967, 89, 3126.See, for esnmplr, D. B. Rorabncher, Inorg. Chenz., 1966, 5,1891.ture-jump methods, are reported for formation in aqueoussolution of complexes between 5-nitrosalicylate(2 -)(nsa, which acts as a bidentate ligand) and hydratednickel(II), and the complexes of nickel(I1) with polytri-phosphate (tp), iminodiacetate (ida), nitrilotriacetate(nta), ethylenediamine-NN’-diacetate (edda), diethylene-triamine (dien), triethylenetetramine (trien), and 2,2’,2”-triaminotriethylamine (tren) .The kinetics are alsoreported for the reaction of pada with the complex[Ni(edda) 3.EXPERIMENTALSolutions were prepared from nickel(i1) nitrate (Fisons,AnalaR) and standardized against the disodium salt ofEDTA. 5-Nitrosalicylic acid (Fluka) and Na,P,O,,, 6H,O(Albright and Wilson) were twice recrystallized from waterand aqueous ethanol, respectively, and 2,2’, 2”-triaminotri-ethylamine (tren) was isolated from technical grade tri-ethylenetetramine (trien) .s The following chemicals wereused without further purification : nitrilotriacetic acid,diethylenetriamine (dien) , trien (Fluka, puriss.grade) ;iminodiacetic acid, 2,4,6-collidine (Fisoiis) ; ethylenedi-amine-NN’-diacetic acid (K. and K.) ; pyridine-2-azo-p-dimethylaniline (pada) (Sigma).The solutions were made up with triply distilled water,the middle distillation being from alkaline potassium per-manganate. An ionic strength of 0 . 3 0 ~ was maintainedwith NaNO, and the solutions were buffered with 2,4,6-collidine (ca. lO-,w). The pH ranges used are quoted inTable 1; no evidence was found for metal hydrolysis inthese regions, although there was curvature in the kineticplot for the complex [Ni(tp)13- (tp 7 polytriphosphate)above pH 8.5.Kinetic and equilibrium measurementswere undertaken by standard methods ‘-lo and in all casespseudo-first-order conditions were maintained (metal con-centrations in the range 10-3-10-1~ being used). Thekinetic data reported here refer to relaxation effects ob-served in the 100 ps-2 s region and were generally obtained5 J. P. Hunt, Co-ovdination Chern. Rev., 1971, 7, 1.D. W. Margerum and H. M. Rosen, J . Amer. Chem. SOC.,1967, 89, 1088; J. P. Jones, E. J. Billo, and D. W. Margerum,ibid., 1970, 92, 1875.M. A. Cobb and D. N. Hague, J.C.S. Fnraday I , 1972, 68,932.8 L. J. Wilson and N. J. Rose, J .Amer. Chem. SOC., 1968, 90,6041.9 G. R. Cayley and D. N. Hague, J.C.S. Faraday I , 1972, 68,2259.10 G. R. Cayley and D. N. Hague, Trans. Favaday SJG., 1971,67, 786250at 380 5-nitrosalicylate(2-) (nsa) or 550 nm (pada), al-though identical relaxation times were obtained at severalother wavelengths. Temperatures were accurate tok0.2 "C.RESULTSThe mechanism postulated for the formation of the coni-plex [Ni(nsa)] from nickel(I1) and 5-nitrosalicylate( 1 -)k**Hnsa + Xi2+ [Ni(Hnsa)]+k,,k..1p.. k4l k l s 1 l k s lk14Hf + nsa + Xi2+ m. [Ni(nsa)] + H+SCHEME 1(Hnsa) is shown in Scheme ( 1). It is a two-path mechanismin which the path involving the fully deprotonated form nsa25rJ.C.S.1 + K ~ c HDalton3.0FIGURE 1974 251I I I I 1 I2.0 4.0 6.0104 ‘INi(tren)l lMI + K ~ c HFIGURE 1is important at higher pH and the other, involving the formin which the phenolic oxygen atom is protonated, is im-portant at lower pH.The relaxation expression for thelonger of the two observed relaxation effects is as in equation(1),9 where K , = k4,/k14, K , = k4,/k,,, cn and c ~ i are thehydrogen- and nickel-ion concentrations, respectively, andactivity coefficients have been neglected. The terms in K,,and A,, could be neglected and k41 was evaluated from a plotof 7-l against c ~ i ( 1 + K2c&l using pK, values for Hnsareported previously (Figure 1).Reactions of nsa with the substituted nickel species[NiL] follow a similar scheme but in these cases it waspossible to evaluate k,, in addition to k41 (the k,, termsalone being negligible).The results are given in Figure 1and Table 1. The [Ni(edda)]-pada system (edda = ethyl-enediamine-NW-diacetate) was analysed by the methoddescribed previously; the results are given in Figure 2 andFIGURE 2 Variation of ~ - 1 with concentration for reaction ofthe complex [Ni(edda)] with pada a t different temperatures.The vertical lines on each point indicate experimental scatterTable 2. Stability constants of most of the complexes weredetermined spectrophotometrically 7, at 25.0 OC. Theagreement between spectrophotometrically and kineticallydetermined values (Tables 1 and 2) is satisfactory.DISCUSSIONScheme (2) illustrates the mechanism of formationof the complex between [Ni(H20)6]2+ and the bidentateligand L-L outlined above. If the steady-state approxi-mation is applied to formation of the intermediate[(H20)5Ni2+-L-L], the observed rate constant for forma-tion of the chelate complex is given by kf = KosK67k78/( k , + k78) which, if k,, $ k,,, reduces to K,-,,k,, and theCaption to Figure IFIGURE 1 Variation of v1 with c ~ ( l + K2c=)-l for reactions ofnickel species with nsa at different temperatures [the verticallines on each point indicate experimental scatter (representing,on average, five or six experiments)] : (a) Ni2+; (b) [Ni(tp)13-;(c) [Ni(ida)] : (d) [Ni(nta)]-; (e) [Ni(edda)] ; (f) [Wi(dien)12+;(9) [Ni (trien)] 2+ ; (h) [Ni(tren)] 2252 J.C.S.Daltonkinetics of formation of the chelate complex are identicalt o those for a unidentate ligand.However, if the firstKO0 [(H,O),NiI2+ + L-L @ (H20),Ni2+(OH2) * - - L-LSCHEME 2metal-ligand bond is very weak (Le. k,, $- k7J theexpression reduces to Kf = K&&8/k76 and the rate-determining step now becomes ring closure.first binding group of the ligand forms a particularlyweak bond to the metal ion (e.g., when it binds throughoxygen ll). Although nsa forms a six-membered ringwith Ni2+(aq) in which both metal-ligand bonds involveoxygen, the value of kf (Table 1 ; log kf = 5-01> is coin-parable t o those found for complex formation betweennickel(I1) and many other dianioiis,l and we concludethat the rate-determining step is replacement of the firstwater molecule in the inner co-ordination sphere of themetal ion (i.e.k78. k7J. This conclusion is strengthenedby the observation that A H $ (13.1 kcal mol-l) is com-parable to activation enthalpies for water exchange atnickel(I1) [for which several values have been quoted, theTABLE 1Rate and equilibrium constants and activation parameters for reaction of nickel(r1) species with nsa (estimated errorsin parentheses)Ni2+(aq) [Ni(tp)13- ( n = 3) [Ni(ida)j ( 1 2 = 3) [Ni(nta)]- (n = 4)7.5-8.0 7.4-8.4 7.5-8.4 7.51*02(0*15) x lo6 1-48(0*15) x lo4 6*3(0*4) x lo4 1.48(0.22) x lo1PHk41a/1 mol-1 s-1 1.0 x 105 3.0 x 10, 1.3 x lo5 4.4 x 104k4Jl mol-l s-l11*4(2-4) 14-3( 1.4) AH,,t/kcal mol-l 13.1 (0.8) 16*0( 1.0)k14/~-' 0.2 2- 9 (0.8) 5*2(1*3) 3 *5 (0.4)AH14*/kcal mol-1 17*1(1*0) 16*6(3.1) 16*3( 1.1)log,, K, (a) kinetic 3*71(0-18) 4*07(O-17) 3.53(0*21)(b) spectrophotometric 5.62 (0- 1 5) 3.38 (0.15) 3.53 (0.1 5)log,, K l e (mean) 5-62 3-85 4.37 4-01AS,,:(s)/cal K-1 mol-l +8(3) + 16(3) +3(7) +11(5)AS,,t/cal K-l mol-l + l(3) -I- 1(9) -1(4)[Ni(edda)] (n = 4) [Ni(dien)12+ ( n = 3)2.6(0.3) x 104 1.53(0*05) x 10, 4*6(0*2) x 10,[Ni(trien)12+ (12 = 4) [Ni(tren)12+ (n = 4)5.2(0.3) x lo67.3-8.2 7.5-8.4 7.5-8.3 7.5-8.5 PI3k41n/l mol-l s-l 7.8 x 104 3.1 x 106 1.4 x 30' 1.6 x 107AS4,*@)/cal I<-1 mol-1 +a@)14 is-' 20(2)ASl,*/cal K-1 mol-l - 4(3) O(5)h4J1 mol-l s-lAH, ,t/kcal mol-l 1 1 a 3 (2.5) 1%5( 1.5) 9*5( 1.8) 8*5(1*3)220(25) 650( 50) 930( 140)14*3( 1.4) 12-2 (2 '4) 11 -6( 1-4)3*85(0*06) 3.75 (0.09)(b) spectrophotometric 3.1 6 (0.08) 3-74(0-14) 3.72 (0-12)+ 13(5) +6(6) +3(4)--5w - 6(5)AH,,:/kcal mol-l 14*3( 1.0)log,, K , (a) kinetic 3.1 l(O.09) 3*84(0*07)log,, Kla (mean) 3.61 4.14 4.28 4.22Rate constants refer to 25 "C and ionic strength 0 .3 ~ (NaNO,) ; n is the number of co-ordination positions of Ni2+ assumed toa Estimated; a superscript s indicates that a statistical correction has been applied (see text).be occupied by the ligand L.It has been suggested that, with NiII, the two stepscan become comparable when the ring-closure step isTABLE 2Rate and equilibrium constants and activation parametersfor reaction of the nickel(I1)-ethy1enediamine-N"-diacetate complex with pada (estimated errors inparen theses)hf/l mol-l s-lAHf:/kcal mol-lASfr(8),/cal K-' mol-lAHdr/kcal mol-lASd:/cal K-l mol-llogl,K (a) kinetickn/s-l( b ) spec tropho tome tricRate constants refer to 25 "C and ionic strength = 0.331(NaNO,).A superscript s indicates that a statistical correctionhas been made to allow for the fact that in the complex[Ni(edda)] there are only two remaining water molecules (seethe text and ref. 7).sterically hindered (especially when the resulting chelatecomplex contains a six-membered ring3) or when thelatest being 12.1 rf 0-5 (ref. 12) and 13.88 kcal rnol-l(ref. 13)] and the reaction of Ni2+(aq) with several other1igands.l~' Thus, for example, with the neutral ligandspada, 1 ,lo-phenanthroline, and 2,2'-bipyridine tlievalues are 1 3 ~ 6 , ~ 1 3 ~ 1 , ~ ~ and 13.2 l4 kcal mol-l, respec-tively, and the difference in kf between nsa and theneutral ligands (a factor of 50--100) merely reflects thegreater desolvation associated with the charge neutralisa-tion which occurs on forming the transition state in theformer case (seen as a more positive AS$ value).Partial replacement of tlie inner hydration sphere ofthe metal ion by a multidentate ligand L might beexpected to affect kl in several ways.(i) A reduction ink f is expected on statistical grounds since the number11 E.g., H. Hoffmann, Brit. Biiiz F~iZgrJrlEschaft Phys. C h ~ n i . ,12 M. Grant, H. W. Dodgen, and J . P. Hunt, J . Azner. CIwii.l3 J. W. Neely and R. E. Connick, J . -qiizer. Chent. SOC., 1072,14 R. H. Holyer, C.D. Hubbard, S. F. ,4. Kettle, and 1%. G.1969, 73, 432.Soc., 1970, 92, 2321.94, 8646.Wilkins, INorg. Chem., 1965, 4, 9291974 253of replaceable water molecules is reduced; this can beallowed for by multiplying the measured k f by anappropriate statistical factor [in the present case by6/(6 - n), where $2 is the number of co-ordination posi-tions occupied by L, since the co-ordination number ofNi2* remains a t six] to give kfs. (ii) Strengthening oftlie metal-ligand bonds could be accompanied by aweakening of the remaining metal-water bonds ; thiswould result in an increase in k,,, and so Rfs, and adecrease in AH+. (iii) The value of k,, could be re-duced, thus allowing ring closure to contribute to therate-determining step (this effect would be expected to1)c more important the larger and more highly chargedare L and L-L).(iu) In the case of a negatively chargedincoming ligand L-L, Kos would be reduced if L was alsocharged since the net positive charge on the metal ionwould be reduced. Effects (iii) and (iv) would both leadto a reduction in kfs. Consideration of the rate constantsfor complex formation between nickel species and nsa,KH,, and pada and for water exchange allows us todecide on the relative importance of factors (ii), (iii), andIn Figure 3 most of these rate constants [suitablyadjusted to take account of factor (i)] are shown forhexa-aquonickel( 11) and the seven substituted nickelspecies. The latter have been ordered along the abscissaatccording to the number of bound nitrogen atoms, asindicated, with the result that the overall charge on thenickel species fluctuates along the series.The generalincrease in kfs found with NH,, pada, and H,O as thenumber of bound aliphatic nitrogen atoms increases is,as expected, found also with nsa [factor (ii)]. The otherprincipal feature of Figure 3 is that, for a given complex;NIL], kps for the charged nsa is always larger than fortlic neutral XH, and pada ligands. The ratio of kfsfor nsa and the neutral ligands varies from 15 to 70 fortlie cationic [NiL] species (Ni2 (as), [Ni(dien)12*, [Ni-(trien)12+, and [Xi(tren)12+) whereas it is generally lessthan 10 for the neutral and anionic [NIL] species([IVi(ida)], [Ni(nta)j-, and [Ni(tp)I3-), being smallest(ca.3) for [Ni(nta)I;-. (The larger difference between]ifq for nsa and pada with the complexes [Ni(trien)12+ and'Ni(tren)12 is probably associated with the contributionof ring closure to the rate-determining step for the latterincoming ligand '.) This is an important result and itconfirms the indications of similar studies with the do(k).and high-spin d5 metal ions Mg2+ (refs. 9 and 15) andMn2+ (ref. 15) where, however, the water-exchange datafor the partially substituted ions are generally notavailable. Thus, the complex-formation rate data canbe rationalized in terms of the normal dissociative modelused for 1 : 1 complexes [Scheme (S)] only if the localcharge density on the metal ion is considered rather thanthe overall charge on the substituted metal ion. In/$$&{60 60 50 50 4 0 3 0 2 0 20atoms 1N 1 N 2N 3N 4N ANFIGURE 3 Statistically adjusted rate constant (log kfa) forreaction of nickel(1x) species with charged and neutral ligands.Sources of data: nsa (A), this work; pada (O), this work andref. 7 ; NH, (a), refs. 6 and 7; H,O (a), ref. 5. In the caseof water exchange, the first-order rate constants quoted inref. 5 have been converted into second-order rate constants bytreating the solvent as a reagent of concentration 55.5~ topermit direct comparison with the other rate constantsparticular, the net charge on the metal species shouldnot be used to calculate KO, from the Fuoss equation.2The dissociation rate constants k d for the [NiL(nsa)jcomplexes on the whole reflect the formation rate con-stants hfs, so there is less variation in Ks than in k f s or Kd.{There is notably less difference in Ks as L is varied thanbetween the values for the complexes [NiL(nsa)] and[Ni(nsa)] .} Similar results have been obtained before,16but this is by no means a universal picture.We are indebted to A. R. White, who did some of thepreliminary work on these systems.[3/768 Received, 11th April, 1973315 D. X. Hague, S. R. Martin, and M. S. Zetter, J.C.S. Faradayl6 M . A. Cobb and D. N. Hague, Trans. Furaday SOC., 1971,I, 1972, 68, 37.67, 3069

ISSN:1477-9226    

DOI:10.1039/DT9740000249

出版商:RSC    

年代:1974

数据来源: RSC