JOURNALOFTHE CHEMICAL SOCIETYDALTON TRANSACTIONSInorganic ChemistryKinetic Studies on the Closure of a Chelate Ring in [(2-Dimethylamino-ethyl)dimethylammonium]- and [(3-Dimethylaminopropyl)dimethyl-ammonium] -t r ic h I o ro plat i n u m ( 11) Com plexesBy Giuliano Annibale, Luciana Maresca, and Lucio Cattalini, lstituto di Chimica Generale ed Inorganica,Giovanni Natile,' Cattedra di Chimica Generale ed Inorganica, FacoltB di Farmacia, UniversitB di Bari, ItalyUniversitA di Venezia, ItalyThe title complexes, [PtCI,( Me4-Hen)] and [PtCI,(Me,-Hpn)] [Me4-Hen = (2-dimethylaminoethyI)dimethyl-ammonium, Me,-Hpn = (3-dimethylaminopropyl)dimethylammonium], with the diamine ligand acting as uni-dentate, have been prepared and the kinetics of closure of the chelate ring studied in aqueous solution.The rateconstants of the reaction [PtCI,(N-N)]- [PtCI,(N-N)] + CI- (N-N = Me4en or Me4pn), measured in basicmedium a t 25 "C, were kfcl = 183 f 6 (Me,en) and 1 .OO f 0.03 s- (Me,pn) ; temperature-variation studies gaveAH1 = 43 f 3 (Me,en) and 57 f 2 kJ mol-l (Me,pn), and AS: = -49 f 4 (Me4en) and -46 f 3 J K - l mol-I(Me,pn). These data are interpreted as a measure of the greater conformational stability of the five- versus six-membered ring. The presence of two methyl substituents on the N atoms causes a 50-1 00-fold rate enhancementwhich is a measure of the so called ' Thorpe-lngold ' or ' gem-dimethyl ' effect. Kinetic measurements at pH < 7are interpreted according to a reaction mechanism in which reversible solvolysis of a cis chloride from the ring-opened species leads to two parallel paths for ring closure.The kinetic constants for acid hydrolysis [ k , = (9.5f 0.5) x 1 0-4 (Me4en) and (5.7 f 0.2) x l 0-4 s- (Me,pn)] and chloride anation of the solvento-species [k-, =0.36 f 0.01 (Me,en) and 0.1 4 f 0.01 dm3 mol- s- (Me4pn)] are in the usual range : the products of the acid-dissociation constant of semidetached (i.e. monoco-ordinated bidentate ligand) diamine and the kinetic constantof ring closing, in the chloro- and in the solvento-species are: Kaclkfc, = (7.8 f 0.4) x lo-' (Me4en) and (3.4f 0.3) x 10- lo (Me,pn), and KaH20kfH20 = (1.03 f 0.05) x 1 0-5 (Me4en) and (1.4 f 0.1 ) X 1 0-9 mol dm-, s-l(Me,pn). From data in acidic and basic media, the acid-dissociation constant was calculated as KaCl = 4.3 x 1 O-B(Me,en) and 3.4 x 10- lo mol dm-, (Me,pn), intermediate between those of the di- and mono-protonated freediamine.THE chelate efiect has been studied for a long time inconnection with equilibrium constants of complexformation and more recently attempts have been madeto investigate the chelate effect from a kinetic point ofview by measuring the rates of ring opening and ringclosing in metal complexes of multidentate ligands.In the case of square-planar complexes, kineticstudies on the closure of five-membered chelate ringshave been performed for tra~z~-[PtCl,(en-HCI),],~ [PtCl,-(bama-HCl)] and [PtC12( t aa-ZHCl)] ,3 [PtCl,( Hbas-HCl)]and [PtCl,(bas*HCl)] [en = 1,Z-diaminoethane, bama= bis(2-aminoethyl)methylamine, taa = tris(2-amino-ethyl)amine, and bas = bis(2-aminoethyl) sulphide].Both the acid-dissociation constant of the semidetached(i.e. monoco-ordinated bidentate ligand) 2-aminoethylgroup and the kinetic constant of ring closing weredetermined.The reverse (ring-opening) reaction couldnot be studied in these systems owing to the high stabilityof the chelate complexes. The opening and closing ofchelate rings of different size were studied kinetically incis-[PtCl,(dmso) (enoHC1) J ,5 cis-[Pt Cl,(dmso) (pn-HCl)],and cis-LPtCl,(dmso) (bn*HCl)] (dmso = dimet hyl sulph-oxide, pn = 1,3-diaminopropane, and bn = 1,4-diarnino-butane), however, with the exception of the bn complex,the experimental data did not allow the separation ofthe acid-dissociation equilibrium constant and the ring-closing kinetic constant.The same limitation alsoapplies to the kinetic study of ring closure in trans-[PtCl,(en*HCl) (NH,)] and trans-[PtCl,(pn*HCl) (KH,)] .7The establishment of an equilibrium between achelated substrate and a species containing a scmi-detached ligand is also the first stage in the displacementof polyamines from reactive palladium(r1) and gold(rr1)species, however this step is usually too fast to befollowed kinetically and only a rough estimation of theproduct of the acid-dissociation equilibrium constant andring-closing kinetic constant can be g i ~ e n . ~ ? ~In this paper we report a complete kinetic study of thering-closing process in [PtCl,(Me,-Hen)] and [PtCl,-(Me,-Hpn)]. The significance of the ring size and ofthe methyl substituents on the N atoms upon the rate ofring closing will be discussed2 J.C.S.DaltonEXPERIMENTALCommercial reagent-grade chemicals were used withoutfurther purification.Pveparations .- Tvadzloro [ (2-dinzet~iylanzinoet~iyZ) dimethyl-arnmonium]~Zat~num(~~). A white solution of [Pt(C,H,)Cl-(Me,en)][ClO,] lo (0.2 g, 0.42 mmol) in concentrated hydro-chloric acid (5 cm3) was kept a t room temperature for somedays during which time it became orange-yellow; then, byslow concentration in a desiccator containing concen-trated sulphuric acid and a beaker of potassium hydroxidepellets, it yielded large reddish crystals of the desiredproduct.These were washed with methanol and air dried.The compound is insoluble in acetone and chlorinatedsolvents, and slightly soluble in water (1;ound.: C, 18.9;H, 4.2; C1, 21.2; N, 5.5. [PtCl,(Me,-Hen)] requires C,18.8; H, 4.15; C1, 20.8; N, 5.5%).Trichloro[ (3-dimethylaminopropyl)tlimethylamnioni~m J -platinum(I1) was prepared in a similar way using [Pt(C,H,)-Cl,(Me,-Hpn)][ClO,] as starting substrate l1 {Found : C,19.2; H, 4.3; C1, 24.3; N, 6.2. LPtCl,(Me,-Hpn)] requiresC, 19.4; 13, 4.4; C1, 24.6; N, 6.5%).Kinetics.-Rate data were obtained spectrophotometric-ally by measuring changes of absorbance with time. First-order rate constants were calculated from plots of ln(Al -A , ) against time, where .Al and A , are absorbances a ttime t and after a t least six half-lives respectively; theseplots were linear for at least four half-lives.Experiments a t constant chloride concentration werecarried out in l-cm quartz cells using a complex concentra-tion of mol drn-,.Experiments a t negligible chlorideconcentration for measuring the rates of acid hydrolysis inthe starting substrate and ring closing in the solvento-species were performed in a 10-cni quartz cell using a com-plex concentration of lo-, mol dmU3, under which conditionsthe equilibrium between chloro- and solvento-species iscompletely shifted towards the solvento-species. The sameprocedure was used to obtain the initial solution of solvento-complex to measure the rate of chloride anation.Fast kinetic measurements in basic media were performedon a Durrum D-110 stopped-flow instrument by rapidmixing of a solution of the reactant substrate in HCl(0.05 mol dm-3) and NaCl (0.05 mol dmd3) with a secondsolution containing NaCl (0.10 mol drn-,) and Na,[PO,](0.10 mol drnp3), the resulting pH being ca. 11.5.Kinetic measurements a t pH (7.2 were performed on aPerkin-Elmer 475 spectrophotometer using HClO, foradjusting the pH to 0.7, phosphoric acid-sodium dihydro-genphosphate buffers for pH values in the range 0.7-3.2,and sodium dihydrogenphosphate-disodium hydrogen-phosphate for pH values in the range 5.5-7.2.TABLE 1Values of kobs.for the reaction [PtCl,(Me,-Hen) (OH,)]+ +C1- - [PtCl,(Me,-Hen)] in water a t I = 0.2 moldm-3 (Na[C10,]) and 25.0 "C[H+l 102[C1-]mol dm-3 mol dm-30.20 0.501 .002.003.335.006.678.3310.010*kIb,.S-10.3600.5080.8651.391.942.543.233.80The pH was measured with a E-500 digital pH meterusing a glass electrode, the instrument being calibrated withan equimolar H,PO,-NaH,PO, buffer solution. The ionicstrength was adjusted to a constant value with sodiumperchlorate.The experimentally determined rate constants, ho~~s./s-l,are reported in Tables 1-5.1'ARLE 2Values of kobs.for the reaction ~f'tCI,(Me,-Hpn) (OH2)] +C1- -+ [PtCI,(Rle,-Hpn)J in water a t I =- 0.3 nioldm-, (Xa[ClO,]) and 25.0 "C[H +I 102[ c 1 -1 1O2krJ,,",mol dm-, mol tlm 510.20TAB0.216 0.9481.54 2.842.31 3.973.00 4.814.62 6.846.67 10.00.0 14.6.E 3Values of kol,s.for the reaction \PtCI,(Me,-Hen)] - [PtCl,(Me,en)] + H+ + C1- in water a t I = 0.2 moldm-3 (Na[ClO,]) and 25.0 "C0.010.100.00 0.050.170.490.871.181.512.240.100.310.500.951.321.322.092.754.374.575.017.418.5111.214.114.50.831.232.092.884.797.599.5515.50.5621.954.938.5512.115.523.20.410.901.392.643.363.394.345.758.048.728.6111.913.216.018.618.41.131.512.223.394.747.389.3914.8TABLE 4Values of hobs. for the reaction [PtCl,(Me,-Hpn)] - [PtCl,(Me,pn)] + Hf + C,l- in water at I = 0.3 moldm-3 (Na[ClO,]) and 25.0 "C[Cl-I 1 0-6[ H +]-l 104kob,.mol dm-3 dm3 mol-' 5-10.00 0.302 3.700.537 5.890.661 7.461.58 12.52.45 17.63.09 17.1982 3TABLE 4 (continued)CCl-I 10-s[H+]-l 1 0 4 , ~ ~ .mol dm-3 dm3 mol-l S-10.105.628.9111.714.514.81.232.092.634.175.005.377.249.7721.221.525.025.726.14.788.120.5715.218.619.724.029.9TABLE 5Values of hobs.and activation parameters for ring-closingreactions measured in basic medium ( I = 0.3 moldmP3)koba. A H r AS1 - - 8, -Complex "C s-l kJ mo1-l J K-l mol-l[PtCl,(Me,en) ] - 16.0 96 43 f 3 -(49 & 5)25.5 19035.5 32425.0 1.0037.2 2.4816.0 0.45 57 2 -(46 f 3)RESULTSKinetics of Acid Hydvolysis a d Chloride R nation.-Operating a t low pH ([H+] = 0.20 mol dm-3) it is possible t osuppress the ring-closing reaction and measure the rate ofsubstitution of a chloride by a water molecule in the startingsubstrate with formation of a solvento-species. Sincechloride ion is released in the reaction it was necessary t ooperate a t a low concentration of complex (d 10-4 mol dm-3)in order to ensure complete hydrolysis; under theseconditions, good linear semilogarithmic plots and first-order kinetics were obtained [equation (I)].Only onekobs. = k1 (1)chloride ion cis to the nitrogen ligand was substituted by awater molecule and this was confirmed by performing thering-closing reaction on the solvento-species and analyzingthe product which was always [PtCl,(N-N)] (N-N = Me4enor Me,pn).The anation reaction was studied by adding sodiumchloride to a solution of the solvento-species and followingthe spectral changes. Using enough chloride ion to ensurepseudo-first-order conditions, good linear semilogarithmicplots were obtained.Plots of kobs. against [Cl-] were linearand had a discrete intercept the value of which agreed, withinthe limits of experimental error, with the rate constants ofacid hydrolysis measured in independent experiments[Figure 1, equation (a)].kobs. = k i f k-,[C1-] (2)The values of the intercepts, k,/s-l, and of the gradients,k-,/dm3 mol-l s-l, were calculated from least-squares fitsand quoted uncertainties (Table 6) are 95% confidencelimits.Kinetics of the Closure of the Five-membered Ring inAcidic Media.-The rate of ring closing in acidic media( 0 . 7 ~ pH d 3.2) of [PtCl,(Me,-Hen)] was a function of thehydrogen- and chloride-ion concentrations.Starting from the solvento-species, cis-[PtCI,(Me,-Hen)(OH,)]+, and in the absence of added chloride, thespectrum changed to that of [PtCl,(Me,en)] a t a rate whichdepended upon the pH of the solution.A plot of hobs.against [H+]-1 was linear and passed through the origin43 -cIw l \$ 2 -102[CI-l / rnol drn-3FIGURE 1 Plot of kob4. against [CI-] for the reaction: [PtCl,-(N-NH)(OH,)]+ + C1- [PtCI,(N-NH)], N-N =: Me,en( a ) and Me,pn (b) ; [H+] = 0.20, I = 0.20 (u) and 0.30 ( b ) rnoldm-, (Na[ClO,]), 25.0 "C[curve (a) of Figure 2, equation (3)]; the slope of tliis plot,KfH,OKalioo/mol dm-3 s-l (the reason for this choice ofsymbolism will appear clear in the Discussion section),derived from a least-squares analysis of the primary data isgiven in Table 6.kobs.= ~ f I I , O ~ a l I , O I H + l - (3)Starting from the chloro-species and in the presence ofadded chloride, the ring-closing process occurred a t a rateTABLE 6Values of kinetic and equilibrium constants measureda t 25.0 "CComplexesConstant [PtCl,(ille,-Hen)] [PtCI,(;ZIe,-Hpn)] ak1/S1 (9.5 r i ~ 0.5) x 10-4 (5.7 3- 0.1) x 10-4k-,/dm3 mol-l s-l 0.36 & 0.01 0.14 1 0.01kfH,OKaHaO/ rnol (1.03 $- 0.05) x (1.4 0.1) xdm-3 s-lKiH,o/mol dm-3 C (4.7 4- 0.5) x 10-7kfcl/s-' 183 & 10 1.00 7 0.05Kacl/mol dm-3 4.3 x 10-9 3.4 Y 10-10kfc,Ka~l/ rnol (7.8 & 0.4) x 10 ' ( 3 . 4 - 0.3) x 10-l'dm-3 s-1@ I = 0.2 mol dm (Na[ClO,]). 1 = 0.3 mol t l m 3 (Na-[CIO,]). Under the experimental contlitions used,K*Hao[H+J-l was negligible and could not be determined.which depended upon the hydrogen- and chloride-ion con-centrations.However, a t constant [Cl-], a plot of bobs.against CH'1-l was not linear and the curvature became lesspronounced as the chloride concentration was increased[curves (b) and (c) of Figure 21. This leads to a reactionmechanism in which the acid hydrolysis and chloride anation,and the ring closing in the chloro- and solvento-species,occur simultaneously (Scheme 1).From this set of first-order reactions, and withou4 J.C.S. Daltonand Kiw.lO(hfH,OKAHaO)-l/~ are the slope and the interceptrespectively (their values, derived from a linear least-squares analysis of the primary data, are reported in Table6).The most simple explanation for the different be-haviour of the hle,en and the Me,pn complex [equations (3)and (6) respectively] is that a t the lower acidity used in theI 1 I I5 10 15 2010-2[H+l -'/dr3 mol-'Plot of hobs. against [H+]-' for t h e reactions [PtCI,-(Me,-Hen)(OH,)]+ [PtCI,(Me,en)] + H+ + H,O (a) and[PtCl,(Me,-Hen)] [PtCl,(Me,en)] + H+ + C1- at [CI-] =0.01 ( b ) a n d 0.10 mol dmW3 (c). 1 = 0.20 mol dm-3 (Na[C10,]),25.0 "CFIGURE 2adopting either the approximation of a pre-equilibriumbetween the chloro- and solvento-species [(8 - y)(cc + P)-lz 01 or that of a steady state for the solvento-species[(a,s)(P + y)-l x 01, equation (4) for the hobs. of the slowerprocess is obtained.12 Using equation (4) and substitutingk3hs.= *((a + P 1- y + 8) - [(a + PI2 + (8 - r), +2(a - P)(S - r)l9 (4)for u, (3, and y the expressions for Kohs. of ( l ) , (2), and (3)respectively, it is possible to derive 8 [equation (5)] by aa ,'Chloro - species' .= Solvento - species'SCHEME 1non-linear least-squares curve-fitting procedure.of hfclKaol/mol dm-3 s-l is reported in Table 6.The value6 = KfaK"cl[H+]-1 (5)It can be shown that under our experimental conditionsneither approximation, of a pre-equilibrium or of a steadystate, leads to an accurate fitting of the experimental data.Kinetics of the Closure of the Six-membered Ring in AcidicMedia.-The ring-closing process in [PtCl,(Me,-Hpn)] wasstudied a t lower acidity (5.5 < pH < 7.2) than that used in[PtCl,(Me,-Hen)] (0.7 < pH < 3.2) in order to havemeasurable rates.Starting from the solvento-species and in the absence ofadded chloride, the dependence of the observed rate con-stant upon [H+]-l was more complicated than the linear re-lationship observed for the Me,en complex [curve (a) ofFigure 31.However, a plot of l/Kob, against [H+] waslinear and had a discrete intercept; this is consistent with arelationship of the form (6) where ( k f ~ , ~ K ~ z ~ ) - l / s dm3 mol-1&be. = ~%T,OK%,O[H+]-~{ 1/(1 -k K'H,O[H+]-')> (6)30 I - I q% 200,4 *10H ' I- ' / d rn rnol- 'Plot of k,,,,,. against [Hi]-' for t h e reactions [I'tCl,-(Me,-Hpn)(OH,)]+ [PtCI,(Me,pn)] + H+ + H,O (a) and[PtCl,(Me,-Hpn)] [PtCI,(Me,pn)] + H+ + C1- at [Cl-] =0.10 mol dm-3 ( b ) .FIGUKE 3I = 0.30 mol ~ I r n - ~ (Na[CIO,J), 25.0 "Clatter case the solvento-species, ci~-[PtCl~(Me,-Hpn) (OH,)]+,undergoes extensive dissociation to give the hydroxo-species cis-[PtCIZ(OH) (Me,-Hpn)], which is unreactivetowards ring closing [equation (7)].K~HI~O[H+]-'cis-[PtCl,(Me,-Hpn) (OH,)]+ .=--cis-[PtC12(OH) (Me4-Hpn) J (7)In order to check the validity of this assumption we haveperformed a kinetic run a t pH 11, under which conditionsthe solvento-species is completely dissociated to the hydroxo-species.The estimated first-order rate constant (affected bya rather large error due to experimental difficulties arisingfrom concurrent solvolysis reactions) was GU. 5 x lo-, s-1and possibly arises from a ring-closing reaction in the sol-vento-species "still in equilibrium with the hydroxo-speciesand/or a ring-closing reaction in the hydroxo-species withdisplacement of the chloride trans to the hydroxyl groupsince we believe that, as usually found, the hydroxyl groupis totally inert towards substitution.Starting from the chloro-species and in the presence ofadded chloride the ring-closing process occurs a t a rate whichdepends upon the hydrogen- and chloride-ion concentration[curve (b) of Figure 31.Using equation (4) and substitutingfor u and y the expressions for hobs. in (1) and (6) respectivelyand for p the expression for equation (3) multiplied by 1/( 1 + Ki~,o[H+]-l) to account for the formation, under theactual experimental conditions, of unreactive hydroxo-species, 6 could be determined by a non-linear least-squarescurve-fitting procedure [equation ( 5 ) ] . The value ofkfclKacl so determined is reported in Table 6.It is to be noted that the calculated values of k f ~ + P o 1982 5agreed within a 20% error with the slope of the quasi-linearcurve (c) of Figure 2 and curve ( b ) of Figure 3 indicatingthat a t [Cl-] = 0.10 mol dmV3 the cbncentration, and thecontribution to the total rate, of the chloro-species arepredominant.Kinetics of the Closuve of tlze Five- and Six-mewbered Ringsin Basic Media.-The ring-closing reactions were alsostudied under conditions where the pH was high enough forall the substrate to be in the form of the unprotonatedring-opened species.These reactions were fast enough torequire the use of a stopped-flow apparatus. To ensurethat the bulk of the substrate was in the form of the chloro-species, the starting complexes were dissolved in HCl (0.05mol dm-3) and NaC1 (0.05 mol dm-3) and the reaction in-itiated by mixing this solution with a solution of NaCl(0.10 mol dm-3) and Ka3[P04] (0.10 mol dm-3), the resultingpH being 11.5. The rate constants were pH independentand their values (each the average from a t least threeindependent determinations) measured a t different tempera-tures are given in Table 5 with the calculated enthalpy andentropy of activation.DIscussIoNFrom the experimental data it is evident that areversible solvolysis of the chloride from the ring-openedspecies lcads to two parallel paths for ring closure; theirdcpendence upon the pH is easily understood on thegrounds that the free end of the diamine must be unpro-tonated in order to bond to platinum, that is the ring-closing step is preceded by a rapid acid-dissociationequilibrium.Rforeover at pK >, 5.5 (used for ring-closing reactions in the Me n complex) the solvento-species undergoes acid dissociation with formation of thehydroxo-species which is unreactive towards ring closing.A straightforward reaction mechanism deduced fromthese data is depicted in Scheme 2.?PN-NH+ c I \ jPtCI' 'CInC I N N\ / PtA \ CI C ISCHEME 2A \C I OH2The rate expression derived therefrom is given byequation (4) where a , (3, y , and 6 are related to the kineticconstants by the following relationships, equations (8)-(11).a = kl/(l + Kacl[H+]-l) (8)p = k-,[C1-]/(1 + KaH,OIH']-l + K'lf,o[H+]-') (9)= ~f~,oKa~,o[H+]-l/(l + K'H,O[H+]-~ +K'H,o[H+]-l) (10)6 = kfclKacl[H+]-'/(l + Kacl[H+]-l) (11)Under the experimental conditions of pH < 3.2, theterms Ka[H+]-l and KiH,o[H+]-l are small compared to 1and therefore equations (8)-(11) become identical to theexperimentally determined equations (1) , (2) , (3), and (5)respectively.On the other hand, for pH values in therange 5.5-7.2 only the term Ka[H+]-l can be neglectedin the denominators of equations (8)-(11) and thereforeequation (10) becomes identical to (6).The rate constants measured in alkaline solutions andin the presence of excess of chloride can be directlyassigned to kfcl, neglecting any possible hydrolysis reac-tion since under these conditions the rate of ring closingis much greater than that of solvolysis.The indepen-dently determined values of kfcl when combined with thevalues of kfc&aC1 obtained from studies in acidic mediaallow the evaluation of Kacl.The first data of Table 6 to be discussed are the rateconstants for acid hydrolysis (kl/s-l) and chloride anation(k-,/dm3 mol-l s-l). If we compare the data for the[PtC1,(Me4-Hen)] and [PtCl,(&4e4-Hpn)] complexes withthose for the analogous [PtCl,(NH,)]- ( k , = 5.6 xs-l and k-, = 4.3 x dm3 mol-1 s-l),13 it appears that,while the equilibrium constants ( k l / k 1 ) are very similarin the three cases (2.6 x and 1.3 x lo-,mol dmW3 respectively) indicating a comparable cisinfluence of the three ligands (Me,-Hen+, Me,-Hpn+, andNH, respectively), the rate constants in the first twocases are a factor of ten higher than the third one.Thecis effect of the three ligands can account for thesedifferences; if we apply to the present case the equationderived for the basicity effect of the cis amine on chloridesubstitution by another amine molecule in cis-[Pt-Cl,(dmso)](amine) complexes {log k = -0.4 PKaamine +C; k = rate constant for chloride substitution, dm3mol-1 s-1 and C = constant),14 a ten-fold rate increase(going from NH, to Me4-Hpn+ and Me4-Hen+) is ex-pected. Therefore, we exclude the existence of a neigh-bouring-group effect of the cationic end of the semi-detached diamine on the rates of these reactions, also onthe grounds that this effect should be different for acidhydrolysis and chloride anat ion.By comparing the rates of ring closing in the chloro-(kfcl) and in the solvento-species ( k f ~ , o ) the effect of theleaving group in these unimolecular processes could bestudied.However, it was possible to_measure directlyonly kfcl and not kfHaO, since in basic medium the solvento-species is converted immediately into the unreactivehydroxo-species. Therefore, we can only compare the4.1 products KfCIK*CI and k$,oKk,o and, since it is un-likely that the difference of only one ligand in the co-ordination sphere of platinum will have a significanteffect on the basicity of the unco-ordinated end of thediamine, we can ascribe to kfcl and k f ~ , 0 the differencesobserved in the products kfclKaCI and k f H , & ' t ~ , ~ .Withthis assumption, the solvento-species appears to be morereactive than the chloro-species by factors of 13 and 4 forthe Me,en and Me,pn complexes respectively. Data onthe leaving-group effect, as measured by the Kh,o :Kfcl ratio, relative to the entry of amines in platinum(I1)substrates are scanty in the literature. A 54-foldgreater substitutional lability was found in [Pt (dien)-(OH,)],' relative to [PtCl(dien)] + [dien = bis(2-amino-ethyl)amine] l5 while a 28-fold difference was foundbetween [Pt(C,H,)Cl,]- and [Pt(C,H,)C1,(OH,)].16 Forring-closing reactions, a kfH,o : kfcl ratio of 40 : 1 wasreported for [PtCl(dmso) (Hen) (0H,)l2+ and [PtCl,(dmso)-(Hen)]+,5 while a value of 2 : 1 can be calculated for[Pd(en)(Hen)(OH2)l3+ and [PdCl(en)(Hen)]2+.8 I tappears, therefore, that the leaving-group effect does notvary much regardless of whether the substitution is abimolecular process or a unimolecular chelate-ring clos-ure,From the data in Table 6 it appears that the Me,encomplex undergoes ring closing much more readily thanthe Me,pn complex, the ratio of kfcl values being ca.2 x lo2 : 1 .One can argue that this difference stemsfrom the different basicity of the two amines, howeverwe can also point out, without going into detail, that thebasicity of the chelating diamine exerts its influenceupon the reaction rate in opposite directions through thecis effect and the entering-group effect ; therefore the netresult will tend to be quite small.Moreover, as hasalready been envisaged, the basicity-related cis effectand the entering group effect of an amine operate mainlythrough the transition state; therefore in chelatingdiamines, since both nitrogens become attached to thesame metal atom, it is reasonable to suggest that thedominant influence on the effective basicity involves theinteractions of the two nitrogens with the metal, thelength of the carbon chain serving only to modify theligand ' bite and the orientation of the N substituents.As a consequence, the effective basicity of the nitrogensin these chelating diamines should remain roughlyconstant and therefore the different rates should reflectmainly the greater conformational stability of the five-veYsZts six-membered rings.17The activation entropies (Table 5 ) are more positivethan those usually found for bimolecular substitution re-actions in platinum(xx) complexes (-60 to -130 J K-lmol-I) l8 and are consistent with an intramolecular process.Moreover, while the entropy is very similar for bothcomplexes, the enthalpy of activation is lower by ca.15 kJ mol-l for the Me,en complex than the Me,pncomplex.Therefore, the greater stability of the five-veysus six-membered ring stems from a more favourableenthalpy rather than a more favourable entropy.lgIt is well established that gem-dimethyl or similarJ.C.S.Daltongroups cause a remarkable stabilization of small rings andalso increase the rate at which small rings are formed.This is called the ' Thorpe-Ingold or ' gem-dimethyl 'effect.20 It would therefore be of interest to observe theextent to which such a steric effect operates when anatom (platinum) much larger than carbon is involved.For this purpose we must compare the rates of ringclosing in unsubstituted and NN-dimethyl substitutedamines. Until now we have been unable to prepare com-plexes with unsubstituted en and pn ligands analogous tothe ones reported in this work; however, we can calcu-late the rate of ring closing in a hypothetical [PtCl,-(L-N)]- complex [L-N being a L-bonded (2-aminoethy1)-L ligand] using the data of related complexes such ast~arzs-[PtCl,(en),],~ [PtCl,(bama)] ,3 and [PtCl,(ba~)]-,~ allhaving an unco-ordinated 2-aminoethyl group.Aftercorrecting for the cis- and trans-effects of non-partici-pating ligands other than chlorideJ21 the values kfa 2.1,3.5, and 3.6 s-l are calculated from the data of the threecomplexes given above.In spite of the gross approximation and the differentnature of L, the agreement among the three calculatedvalues is fairly good. If we compare these with thevalue of 183 s-l measured for the Me,en complex we mustconclude that the ' gem-dimethyl groups cause aremarkable 50-100-fold increase in the rate constant.A similar conclusion can be drawn from comparison of' effective molarity ' in unsubstituted and NN-dimethyl-substituted 2-aminoethyl complexes.The ' effectivemolarity ' of the free end of a monoco-ordinated biden-tate ligand has been defined as the ratio between thefirst-order rate constant for ring closing and the second-order rate constant for chloride-ion substitution byammonia in a strictly related complex. Without cor-recting for any basicity effect, an ' effective rnolarity 'of 1 x lo3-4 x lo3 mol dm-3 is always calculated for anunsubst ituted 2-aminoethyl group in platinum complexes.On the other hand, the ratio between the rate of ringclosing in LPtCl,(Me,en)]- and that of cis-chloride sub-stitution by ammonia in [PtC13(NH3)]- 22 is ca. lo5 moldm-3 which is 50-100 times higher than the usualvalue. This gives further support to our statement.I t is finally to be noted that the acid-dissociation con-stant of semidetached diamine, Kacl, is intermediatebetween those of di- and mono-protonated free diamine 23although the actual value is closer to the value for thelatter.We also note that the acid-dissociation constantof the solvento-species to give the hydroxo-species ,KiIr,o, is similar to that observed in cis-[PtCl,(OH,)-(XH,)] (Ki = lop7 mol dm-3).24We are grateful to Dr. G. Tauzher for assistance in kineticmeasurements and to Consiglio Nazionale delle Ricerche(C.N.R.), Roma, for financial support.[1/507 Received, 30th March, 19811REFERENCES1971, 168, 427.1 For a fairly recent review see G. Anderegg, ACS Monogr.1982M. G. Carter and J . K.Beattie, Inorg. Chem., 1970, 9, 1233.G. Natile, G. Albertin, E. Bordignon, and A. A. Orio, J .Chem. SOC., Dalton Trans., 1976, 626.* G. Albertin, E. Bordignon, A. A. Orib, B. Pavoni, and H. B.Gray, Inorg. Chem., 1979, 18, 1451.R. Romeo, S. Lanza, and M. L. Tobe, Inovg. Chem., 1977, 16,785.13 R. Romeo, S. Lanza, D. Minniti, and M. L. Tobe, Inorg.Chem., 1978, 9 , 2436.0. Monsted and J . Bierrum, ‘ Progress in Co-ordinationChemistry, Proceedings of the XIth International Conference onCo-ordination Chemistry,’ ed. &I. Cais, Elsevier, Amsterdam, 1968,J . S. Coe, J. li. Lyons, and M. D. Hussain, J . Chem. SOL. A ,G. Annibale, G. Natile, B. Pitteri, and L. Cattalini, J . Chem.lo L. Maresca, G. Natile, and G. Rizzardi, Inoyg. Chim. A d a ,p. 103.1970, 90.S O C . , Daltopf ?‘i,atzs., 1978, 728; and refs. therein.1980. 38, 137.I,. hfaresca, unpublished work.l 2 A . A. Frost and I<. G. Pearson, ‘ Kinetics and Mechanisms,’2ntl edn.. JViley, New York, 1961, pp. 173-177.l3 Rl. ,I. Tucker, C. B. Colvin, and D. S. Martin, jun., Iworg.Chem., 1964, 3, 1373; I). S. Martin, Inovg. Chim. A d a Rev., 1967,87.l4 P. D. Braddock, R. Romeo, and M. L. Tobe, Inorg. Chem.,1974, 13, 1170.15 L. Cattalini, ‘ Reaction Mechanisms in Inorganic Chemistry,’Inorg. Chem. Series One, Butterworths, Oxford, 1972, vol. 9, ch. 7.16 G. Carturan, P. Uguagliati, and U. Belluco, Inorg. Chem.,1974, 13, 542.l7 E. J . Corey and J . C. Bailar, jun., J . A m . Chem. Soc., 1959,81, 2620.18 U. Belluco, R. Ettorre, F. Basolo, R. G. Pearson, and A.Tutco, Inorg. Chem., 1966, 5, 591.l9 F. Basolo and R. G. Pearson, ‘Mechanisms of InorganicReactions,’ 2nd edn., Wiley, New York, 1967, p. 29.2o E. L. Eliel, ‘ Stereochemistry of Carbon Compounds,’hfcGraw-Hill, New York, 1962, pp. 197-202; E. L. Eliel, N. L.Allinger, S. J . Angyal, and G. A. Morrison, ‘ ConiormationalAnalysis,’ Interscience, New York, 1965.21 L. I. Elding and 0. Groning, Inorg. Chem., 1978, 7, 1872.22 C. B. Colvin, R. G. Gunther, L. D. Hunter, J. A. McLean,M. A. Tucker, and D. S. Martin, jun., Inorg. Chim. Acta, 1968, 2,487.23 T. S. Turan and D. B. Rorabacher, Inorg. Chem., 1972, 11,288.24 T. S. Elleman, J. W. Reishus, and D. S. Martin, jun., J . Am.Chem. SOC., 1958, 80, 536