JOURNALOFTHE CHEMICAL SOCIETYDALTON TRANSACT1 0 N SInorganic ChemistryStructural and Mechanistic Studies of Co-ordination Compounds. PartV111.substituted Tetra-aminecobalt( 111) CationsPreparation, Aquation, and Base Hydrolysis of Some trans-Mono-By Chung-Kwong Poon * and Ha-Wai Tong, Department of Chemistry, University of Hong Kong, PokfulamRoad, Hong KongThe preparation and characterization of trans-[Co(NH,),CNBr]+, trans-[Co(cyclam)CNBr]+, and trans-[Co-(NH,),N,CI] + cations and an improved preparation of trans- [Co(cyclam) N,CI]+ cation are described. The aqua-tion and base hydrolysis of these complex cations have been studied over a range of temperature. The thermo-dynamic and kinetic data are discussed, as before, in terms of the nephelauxetic effect of the amine ligands on thecentral cobalt(ll1) ion.Kinetic evidence i s produced to support a dissociative mechanism for the base hydrolysis ofoctahedral cobalt( Ill)-amine complexes.FOR a given pair of unidentate ligands A and X of com-plexes of the type trans-[Co(am),AX]n+, the difference inthe thermodynamic and kinetic properties for (am), =cyclam, (en),, and (NH,),, where cyclam represents1,4,8,1 l-tetra-azacyclotetradecane and en representsethylenediamine, is governed primarily by the nephel-auxetic effect of the amine ligands on the centralcobalt (111) Thermodynamically, the relativestability of a lower-charged complex to a higher-chargedcomplex is expected to decrease along the above series of(am),. Kinetically, the aquation rate constants of theseanalogous complexes would increase but the basehydrolysis rate constants decrease along the same series.These suggestions were found true for an extensive seriesof chloro-complexes with A = C1-, NCS-, NO,-, CN--,and NH,.For complexes containing a different leavinggroup X, only the trans-[Co(am),NO,Br] + system hasbeen and it was found that this system alsoobeyed the above suggestions. It is, however, desirableto investigate another bromo-system to see if thesuggestions are of general validity. We also intend toextend the study of the chloro-series to include azide asorienting ligand. This paper describes the preparationand kinetic studies of trans-[CoLCNBr] + and trans-.[CoLN,Cl]+, where L = (NHJ4 and cyclam.ThePart VII, W. I<. Lee and C. K. Poon, Inorg. Chem., 1973,12, 2016.C. K. Poon, .J. Amer. Chem. SOC., 1970, 92, 4467.C. K. Poon, Inorg. China. Acta Rev., 1970, 4, 123. * C . K. Poon, Co-ordination Chem. Rev., 1973, 10, 1.C. K. Poon and H. W. Tong, J . Chem. Soc. ( A ) , 1971, 2161.C. K. Lui and C. I<. Poon, J.C.S. Dalton, 1972, 216.corresponding rate data of bisethylenediamine analoguesare known.8-1°EXPERIMENTALtrans-Azidoaquo( 1,4,8,1 l-tetra-azacyclotetradecane) cobabt-(111) Perchlorate.-Perchloric acid (5 ml; 70%) was added t oa suspension of y-peroxo-bisitrans-azido( 1,4,8,1 l-tetra-aza-cyclotetradecane)cobalt(m) )perchlorate l1 in water (2 g in10 ml). The solution was heated on a steam-bath for 10 minuntil a clear blue solution was formed. Excess of alcoholand ether were added successively to precipitate the product[Co(C,,H,,N,)N,H,O] (ClO,), which was filtered off, washedwith absolute alcohol and ether, and dried (PzO,) (yield1-2 g) (Found: C, 23.4; H, 4-95; N, 19.2; C1, 14.0.Calc.for C,,H,,Cl,CoN,O,: C, 23.0; H, 5-05; N, 19.1; C1,13.7%).trans-Chloroazido ( 1,4,8,11-tetra-azacyclotetradecane) cobalt-(111) Perchlorate.-Bosnich, Poon, and Tobe prepared this inpoor yield.12 An improved method involved the prepar-ation of the intermediate trans-[Co(cyclam)N,OHz] (C104)2,to a saturated solution (2 g in 10 ml) of which was addedconcentrated hydrochloric acid (2 ml) dropwise with stirring.The solution was allowed to stand in a salt-ice bath for + h.Bluish green crystals of trans-[Co(cyclam)N,Cl]ClO, beganto separate, which were filtered off and recrystallized fromC.H. Langford and M. L. Tobe, J . Chem. SOC., 1963, 506. * S. C. Chan, J . Chem. SOC., 1964, 2716.P. J. Staples and M. L. Tobe, J . Chem. SOG., 1960, 4803.lo V. Ricevuto and M. L. Tobe, Inorg. Chem., 1970, 9, 1785.l1 B. Bosnich, C. K. Poon, andM. L. Tobe, Inorg. Chem., 1966,l2 B. Bosnich, C. K. Poon, and M. L. Tobe, Inorg. Chem.,6, 1514.1965, 4, 11022 J.C.S. Daltonboiling dilute perchloric acid (ca. ~O-,M). The product wascollected, washed with alcohol and ether, and air-dried(yield 0.8 g) (Found: C, 23.5; H, 4.9; N, 19-2; C1, 14.0.Calc. for C,,H,,Cl,CoN,O,: C, 23.2; H, 5.1; N, 19.1;trans-Chloroazidotetra-amwinecobaZt(m) PercMorate,-trans-[Co(NH,),Cl,]N, (10 g), prepared by the rapid additionof a cold saturated aqueous solution of trans-[Co(NH,),Cl,]-C1 l3 to an ice-cold concentrated solution of sodium azide,was suspended in a boiling aqueous solution, saturated withNH,Cl, until most of the solid had gone into the solutionwhich gradually turned purple. The solution was filteredhot and the filtrate was cooled in an ice-bath. Excess ofmethanol was added to the filtrate to precipitate most of theundesired products which could be shown to containNH4Cl, the unchanged dichloro-complex, and cis-chloro-azido- and other diazido-complexes.The desired trans-chloroazido-complex was most soluble and remained insolution. Ether was then cautiously added to the filtrateuntil the solution appeared cloudy and trans-[Co(NH,),-N,Cl]Cl slowly precipitated, contaminated by hTH4Cl.Itwas recrystallized three times by cautious addition of etherto a saturated methanolic solution. The purified compoundwas then dissolved in a minimum of methanol and concen-trated perchloric acid (2 ml; 70%) was added. Addition ofether precipitated the desired tvans- [Co (NH,),N,Cl]ClO,(yield, 0.5 g) (Found: H, 3-80; N, 32-7; C1, 23.8. Calc. forH,,Cl,CoN,O,: H, 3.98; N, 32.3; C1, 23.3%). The pooryield is the consequence of the need to obtain an analyticallypure sample. Much higher yields can be obtained with5-10% contamination by NH,Cl.trans-Bromocyunotetra-anzl.ninecobaZt(~~~) Bromide.-trans-[Co(NH3),CNOH,]C1, l4 (2 g) was suspended in boilingbromine-free hydrobromic acid (50 ml; 48% ; redistilledfrom red phosphorus).The solid gradually dissolved. Thesolution, after being concentrated to ca. 20 ml, was filteredhot. On cooling, yellowish orange crystals of tvans-[Co(NH,),CNBr]Br slowly separated. These were filteredoff, washed with ethanol and ether, and dried in a vacuumdesiccator (yield: 1 g) (Found: C, 3-70; H, 3.90; N, 22.2;Br, 50.4. Calc. for CHl,Br,CoN5: C, 3-84; H, 3-87; N,22.4; Br, 51.0y0). Attempts to recrystallize the compoundfrom aqueous solution resulted in the formation of thecyanoaquo-compound. This compound was not sufficientlysoluble in most organic solvents t o make recrystallizationpossible. For the same reasons, attempts t o prepare adifferent salt of this complex cation have not been successful.C O ~ U Z ~ ( I I I ) Nitrate.-The method of preparation of thiscomfiound [Co(Cl,H,,N,)CNBr~NO,, followed closely that ofthe ~hloro-analogue,l5~1~ from bromine-free hydrobromicacid instead of hydrochloric acid (yield : 40%) (Found :C, 31-2; H, 5-60; N, 19.8; Br, 19.0.Calc. for C11H24-BrCoN,O,: C, 30.9; H, 5.66; N, 19.7; Br, 18.7'30). Theyield can be improved by addition of alcohol and etherto the filtrate to precipitate a second crop.Kinetics.-All these reactions, aquation and base hydroly-sis, were followed spectrophotometrically in situ with use ofeither a Unicam SP 700 or SP 8000 recording spectrophoto-meter in a conventional manner as described previously.6* l6C1, 13.7%).trans-Bromocyano( 1,4,8,11 -tetra-azacyclotetradecane) -l3 S. M.Jorgensen, 2. anorg. Chem., 1897, 14, 404; ' Hand-book of Preparative Inorganic Chemistry,' vol. 2, 2nd edn.,ed. G. Brauer, Academic Press, New York, 1965, p. 1537.l4 . B. Baranovskii and A. V. Babaeva, Russian J . Inorg.Chem., 1964, 9, 1168.RESULTSThe behaviour of trans-[Co(cyclam)N,Cl]+ and trans-[Co(cyclam)CNBr]+ in dilute nitric acid (0.01-0.1~) wasvery similar to that of other cyclam complexes of the sametype, trans-[Co (cy~lam)AX]+.~. l 6 2 l 6 The changing visiblespectrum maintained isosbestic points throughout the entirereaction (at 584 nm for the azicto-complex and a t 400 and467 nm for the cyano-complex). These isosbestic pointswere crossed by the spectrum oi the corresponding trans-[C~(cyclam)A(OH,)]~+, where ,4 = N,- and CN-- respect-ively. The initial spectrum was identical with that of thestarting complex.Volhard's titration confirmed that theco-ordinated halide was released to cn. 60 and 90%respectively. Addition of an excess of the correspondinghalide to the final solution forced the reaction to re-trace itsown path. These observations clearly demonstrated thatthese cyclam complexes only partially aquated with com-plete retention of configuration. The reaction was followedspectrophotometrically a t 550 and 350 nm respectively.The concentration X of the aquo-complex generated a t anytime t was determined from the observed absorbance at thattime by use of the appropriate known molar absorptivities(E for S,C1 = 118 and for N,OH, = 250 1 mol-l cn1-l at 550nm and for CNBr = 149 and CNOH, = 78.9 1 m01-l cm-1 at350 nin).The forward aquation rate constant, K,, was ob-tained from thc slope of the linear plot of the left-hand partof expression (1) 1' against time, where n = [startingcomplcx] at t = 0, X, = laquo-complex! = [halide] a tequilibrium. These plots were straight to two half-lives.The behaviour of trans-[Co(NH,),CNBr]Br in 0-liu-nitricacid was similar to that described above except that here therelease of bromide was complete. The changing spectrummaintained isosbestic points at 399 and 457 nm with thefinal spectrum identical with that of trans-[Co(NH,),-CNOH,I2+. The reaction was followed a t 500 nm and thefirst-order rate constant was obtained from the standardsemilogarithmic plot of log (Dt - Dm) against time, whereDt and 23, are respectively the absorbances at time t andafter I0 half-lives.The spectrophotometric change associated with theaquation of trans-[Co(NH,),N,C1]ClO4 in 0-OlM-nitric acidinitially maintained isosbestic points a t 448 and 562 nni.The initial spectrum was identical with that of the startingtvans-chloroazido-complex.Volhard's titration confirmedthat chloride was gradually released from co-ordination .At a later stage of the reaction, the changing spectra beganto deviate from these isosbestic points while a new isosbesticpoint was gradually developed at 546 nm which was main-tained until the end of the reaction. Colorimetric azidcdetermination l8 confirmed that azide was partially releasedfrom co-ordination during the second stags of the reaction toca.60% a t equilibrium. Since no attempt to prepare purecis- and tvans-[C~(hrH,),N,OH,]~~ cations or to obtain theirabsorption spectra by any indirect method has been success-ful, the steric course of the aquation of tva?~,s-[Co(NH,),N,Cl]~was deduced by an indirect method. The Hg2"-induceclreaction of this complex cation was followed at a relatively15 K. S. Mok and C. K. Poon, Inorg. Chem., 1971, 10, 226.16 C . K. Poon and H. W. Tong, J.,C.S. Dalton, 1973, 1301.17 A. A. Frost and R. G. Pearson,2nd edn., Wiley, New York, 1961, p. 187.18 P. J. Staples, Chem. and Industry, 1960, 1210.These plots were linear to 3 half-lives.Kinetics and Mechanisms,1974 3low Hg2+ concentration and it was found that the associatedspectral change was virtually identical with that of thespontaneous aquation.In the presence of a large excess ofHg2+, the induced reaction was virtually complete as soon asthe complex was dissolved in the reagent solution. Thevisible spectrum then immediately measured had an absorp-tion peak a t 540 nm and the manner of the subsequentspectral change was the same as that of the second step inthe spontaneous aquation. These observations seemed tosuggest that the steric courses for both spontaneous andI-Ig2+-induced aqu ation of trans- [Co (NH,) ,N3C1]+ wereidentical. With reference t o the well known stereoretentiveaquation of trans-[Co(en),N,Cl]+ lo and HgZ+-induced aqua-tion of most halogonocobalt(II1)-amine complexes , and tothe known spectra of cis- and trans-[Co(en) 2N,0H2]2+ (A,,at 505 and 550 nm respectively 19), it seems highly likely thatthe aquation of trans-[Co(NH,),N,Cl]+ is steroretentive.'This deduction becomes invalid if the geometrical isomeriz-ation of the azidoaquo-complexes is too fast, even muchfaster than the Hg2+-induced reaction, to be detected.This,TABLE 1First-order rate constants for the aquation oftmns-[Co(am) ,N3C1]C10,(am) 4(NH3) 4(NH3)4(NH3)4(NH3)4VHJ4(NHJ 4(NH3)4(NH,),(NHJ,"H3) 4(NH3)4(NH3) 1"H3) 4(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(cyclam)(NH3)4(en) z(cyclam)t/"C16.718.418.920.021.822.124.525-125.225-729-829.930.259.760-860.962-764.470.672-272.472.772.775- 776.276.977-778.025.025.025.0103~,/~-1 a1-331.661-752.062.522- 633.493.883-774-016.536.676.840.2220.2520.2520-3130.3760.7420.8880.9060.9350.9351.291.351-451-571.633.7 b0.26 "0.0026 d[Co(NH,),N,Cl]+ was not affected by the subsequent reac-tion.The first-order rate constant was obtained from theslope of the standard semi-logarithmic plot. These plotswere linear to three half-lives. All these aquation data arein Tables 1 and 2.TABLE 2First-order aquation rate constants of complexes of thetype trans-[Co(am),CNBr]+(am) 4 t/"C k,/s-l a23-2 (3-30 & 0.02) x34.547.6(1-21 & 0.03) x lo-'(4.60 f 0.15) x lo-'(c yclam) 41.9 (2.39 f 0-05) x lo-'(c yclam) 49.5 (5.83 f 0.10) x(cyclam) 58.2 (1.58 f.0.03) x lo-'(NH3) 4(NH3) 4WH3) 425.0 4.2 x 10-3b25.0 4.7 x 10-4"(NH314(en) 2(cyclam) 25.0 2.8 xa This work except as indicated; each entry represents anaverage of three independent determinations in 0-1~-nitricacid and over a starting complex concentration between 1.2and 2.8 mM. Obtained by extrapolation; also AHS = 19-7 50.3 kcal mol-l, A S = - 3 f 1 cal mol-l K-l. In neutralaqueous solution; S. C. Chan, J . Chew. Soc., 1964, 2716.d Obtained by extrapolation; also AH$ = 23.3 f 0-3 kcalmol-l, ASS = -6 & 2 cal mol-l K-l.The base hydrolysis of these complexes was most con-veniently followed in buffer solutions (for conditions, seeTable 3).It has been shown 2o that the buffers used werenon-interfering. In most cases, the changing visiblespectrum maintained a set of isosbestic points (at 472 and561 nm for trans-[Co(cyclarn)N,Cl]+; a t 376 and 496 nm fortrans-[Co(cyclam)CNBr]+ ; and a t 558 nm for trans-[Co(NH,),N,Cl]') throughout the entire reaction. In thecase of trans-[Co(NH,),CNBr]+, the changing visiblespectrum slowly increased its absorbance without establish-ing any isosbestic point. In every case, Volhard's titrationconfirmed that the release of halide was complete and theabsorption spectrum of the acidified final solution wasidentical with that of the corresponding aquo-complex. Itcan be concluded that the base hydrolysis of these com-plexes is complete and stereoretentive.All these reactionswere studied in a conventional manner by following thechanging absorbance a t a single wavelength (at 600 nm fortrans-[Co(cyclam)N,Cl]+ ; a t 450 nm for trans-[Co(cyclam)-CNBr]+; a t 590 nm for trans-[Co(NH,),N,Cl]+; and a t 350nm for trans-[Co(NH,),CNBr]+). These data are in Table 3.For the relatively slowest base hydrolysis of trans-[Co(NH,),-N,Cl]+, the plots of the pseudo-first-order rate constants,bobs, against hydroxide ion concentration were linear withnon-zero intercepts according to the expression : hobs =0 This work except as indicated; [complex] between 0.9and 1.3 mM; solvent = 0-Oh-nitric acid.Obtained by-1 f 0.5 cal mol-] K-l. In dilute perchloric acid ofunspecified concentration ; V. Ricevuto and M. L. Tobe,Inorg. Chem., 1970, 9, 1785. Obtained by extrapolation;also AH$ = 24.6 i 0.2 kcal mol-l, ASS = -2 5 1 cal mol-'K-1.k , + k2[OH-]. it was further found (Table 3) t h a t t h evalues of k,, which corresponded to the background aquationpolating the aquation rate data determined independently.AH' I= 20*5 f Oe2 kcal mol-', ASS = rate constants, agreed weal with those obtained by extra-DISCUSSIONhowever, is highly unlikely. The conclusion must still beregarded as tentative until the true spectra of cis- and trans-[CO(NH,),N,OH,]~ are obtained and the aquation stericcourse can then be accurately determined.The reactionwas followed spectrophotometrically a t the second isosbesticpoint (546 nm) a t which the study of the aquation of trans-The assignment of a trans-configuration to the newcomplexes prepared was made on the basis of variousmethods, depending on individual circumstances. It wasl9 Estimated from the published spectra of cis- and trans-[Co(en),N30H212+; D. A. Buckingham, I. I . Olsen, and A. M.Sargeson, Inorg. Chem., 1967, 6, 1807.2o C. K. Poon and M. L. Tobe, J . Chem. Soc. ( A ) , 1967, 20694 J.C.S. DaltonTABLE 3Second-order base hydrolysis rate constants ofcomplexes of the type trans-[Co(am),AX]+ a103k,/(am), A X t/"C k,/l mol-1 s-1 S-1(NH3)*; N, C1 13.0 (5-23 f 0.05) x 0.85(NH,), N, C1 18.1 (1.10 rf 0.02) x 10-1 1.6(NH,),b N, C1 24.0 (2-63 f 0.03) x 10-1 3.3(NH,),b N, C1 28.1 (4.55 f 0.04) x 10-1 5.4(NH,), CN Br 13.6 (7.33 f 0.10)(NH,), CN Br 21.9 (2.20 f 0.10) x 10(NH,), CN Br 27.2 (4.56 f 0.06) x 10(cyclam) N, C1 25.4 (2.88 f 0.05) x lo3(cyclam) N, C1 30.2 (5.54 f 0.10) x lo3(cyclam) N, C1 34.5 (9.47 f 0.15) x lo3(cyclam) N, C1 40.2 (1.96 f 0.03) x lo4(cyclam) CN Br 17.5 (2.12 f 0.05) x lo2(cyclam) CN Br 25.4 (4.96 f 0.10) x 102(cyclam) CN Br 34.1 (1.45 0.02) x lo3(cyclam) CN Br 43.4 (4.05 & 0.08) x lo3(en)2 N, C1 0.0 4.1 x 1 0 - l d l e(cyclam) N, C1 0.0 6-8 x 102(NH,), CN Br 0.0 9.9 x 10-lo(en)2 CN Br 0.0 1.8dsh(cyclam) CN Br 0.0 2-0 x 10'(NH,), N, C1 0.0 6.6 x103k1/S-10.841.63.35.4a This work except as indicated; reactions of cyclamcomplexes were studied in y-collidine-nitric acid buffersolutions (pH = 7*0-8-6) and those of tetra-ammine com-plexes in 2,6-dimethylpiperidine-nitric acid buff er solutions(pH = 10.3-12.0) ; I = 0 .1 ~ (NaNO,) ; [complex] between0.9 and 2.8 mM; each entry was obtained from 5 to 7 differentruns over a span of 0.9-1.0 pH unit. b hobe = k , + &[OH-];k, wasTobtained by extrapolation of data in Table 1. C Ob-tained by extrapolation: also AH3 = 24.0 f 0.2 kcal mol-1,AS$ = 20 f 3 cal mol-1 K-l. d Reactions were studied indilute NaOH solution; ionic strength was not maintainedconstant with any supporting electrolyte. 6 P. J. Staplesand M. L. Tobe, J . Chem. SOC., 1960, 4803. f Obtained byextrapolation; also AH* = 23.5 f 0.2 kcal mol-l, ASf =36 f 5 cal mol-1 K-l.B Obtained by extrapolation; alsoAHS = 22.3 rf 0.3 kcal mol-l, ASS = 24 f 4 cal mol-l K-l.h S. C. Chan, J . Chem. SOL, 1964, 2716. i Obtained byextrapolation; also AH$ = 20.1 f 0.4 kcal mol-l, AS* =21 5 5 cal mol-l K-l.confirmed that the most consistent variation between thei.r. spectra of cis- and trans-isomers of cobalt (m)-cyclamcomplexes was in the 800-900 cm-l region.21 Here, thepresence of two bands (at 888 and 900 cm-l) near 900 cm-land only one band (at 824 cm-l) near 800 cm-l in thisregion of the i.r. spectrum of the new [Co(cyclam)CNBr]-NO, complex was taken to indicate a trams-configuration.The assignment of a trans-configuration to [CO(NH,)~-N,C1]C104 prepared was made by comparing its absorp-tion spectrum (A-, = 584 nm) with those of the knowncis- and trans-[Co(en),N,Cl]+ (A,,, = 530 and 584 nmrespectively 22) and trans-[Co(cyclam)N,Cl] + (A,,, =586 nm 12).[Co(NH,),CNBr]Br was assigned a trans-configuration by chemical means in essentially the sameway as that foi the corresponding t~ans-[Co(NH,),-CNCl] C1.1621 C . K. Poon, Inovg. Clzim. Acta, 1971, 5, 322.22 Estimated from the published spectra of cis- and trans-[Co(en),N,Cl]+; P. J. Staples and M. L. Tobe, J . Chem. Soc.,1960, 4812.23 M. L. Tobe, Inorg. Chem., 1968, 7, 1260.The aquation and base hydrolysis rate constants ofthese complexes, extrapolated to 25 and 0 "C respect-ively, are in Tables 1-3. The variation of the aquationand base hydrolysis rate constants with the nature of theamine ligands fully supports the earlier suggestion con-cerning the influence of kinetic nephelauxetic effect onthe lability of octahedral cobalt (111)-amine complexes.Thermodynamically, the relative stability of a lower-charged complex to a higher-charged complex alsodecreases, as predicted, along the series of (am),:cyclam > (en), > (NH,),.Here, trans-[Co(cyclam)-N3C1]C10, aquates to ca. 60% at equilibrium at 25 "C;the aquation of the bisethylenediamine complex isvirtually complete but further displacement of the azido-group is not noticeable; 10 for the tetra-ammine complex,not only is the release of the chloride complete, but therelease of the co-ordinated azide also occurs to about 60%giving a mixture of cis- and trans-[C0(NH,),(0H,),]~+.Similarly for the bromocyano-system, the extent ofaquation of trans-[Co(en),CNBr]Br is ca.97% complete.8This is greater than that of trans-[Co(cyclam)CNBr]Br(ca. 80%) but smaller than that of trans-[Co(NH,),-CNBrIBr. The aquation of the tetra-ammine complexis complete even in the presence of 0-hi-bromide ion.The observed negative entropies of activation for theaquation of these complexes suggest that Tobe's cor-relation m of square pyramidal intermediate and reten-tion of configuration with lower entropies of activation isalso applicable for these bromo-complexes.A search of existing kinetic data shows that changingthe nature of the leaving group from chloride to bromideincreases the rates of both acid and base hydrolysis ofanionocobalt (111)-amine complexes.These data, extra-polated to the same temperature, for an extensive classof complexes containing orienting ligands of differentelectronic displacement property, are in Tables 4 and 5.With the exception of isothiocyanato-complexes, thisleaving-group effect, as measured bythekiBr/kiG1 (a = 1 or2)in both Tables, is significantly greater on base hydrolysisthan on acid hydrolysis. These ratios are not greatlydependent on temperature since the activation energyof a given reaction does not vary much from chloro- tothe corresponding bromo-complex.24~25 This leaving-group effect is fully consistent with the proposed SNlcbmechanism for the base hydrolysis and a dissociativeinterchange mechanism (Id) for the aquation of these com-plexes.3~4,26~27 On the basis of an SNlcb mechanism, thesecond-order rate constant is directly proportional toK, x kcb where K, is the acid ionization constant (amineproton) of the conjugate acid and kcb is the aquation rateconstant of the amido-conjugate base.It is expected 26that a change of the leaving group from chloride to bro-mide would not significantly alter the value of K, and,therefore, the kinetic ratio for base hydrolysis essentially24 S. C. Chan and RI. L. Tobe, J . Chem. SOC., 1962, 4531.25 S. C. Chan and M. L. Tobe, J . Chem. SOC., 1963, 5700.26 F. Basolo and R. G. Pearson, ' Mechanisms of Inorganic27 C. H. Langford and H. B. Gray, 'Ligand SubstitutionReactions,' 2nd edn., Wiley, New York, 1967.Processes,' Benjamin, New York, 19661974 5gives the ratio of the kcb for the bromo- and the correspond- molecular mechanism for the base hydrolyis of theseing chloro-complex.The presence of a good labilizing cobalt (III)-amine complexes. The reason behind theamido-group may promote a limiting S N 1 acid hydrolysis reversed behaviour of the isothiocyanato-complexes isof the conjugate base which would probably be more not clear. Langford and Gray27 have pointed out thatTABLE 4Effect of leaving group on the acid and base hydrolysis of complexes of the type cis- and tmns-CoLAX+ (X = C1and Br) at 0.0 "C QAtrans-NO,trans-NO,trans-CNtrans-CNtrans-CNtrans-C1cis-c1trans-OHc~s-OHtrans-NCSIvans-NO,trans-NCSklBr/klC12.93-95-15.26.53.21.86-95.2131010k 2Br/k 2°(5.08.5312814253.24.77.35.95.310pn = Propylenediamine.6 klBr and k2Br Represent the first-order acid hydrolysis rate constant and the second-order basehydrolysis rate constant respectively of the bromo-complex ; similarly, klC1 and k2a represent those of the chloro-complex. Extra-polated from published data a t other temperatures. At I = 0 . 3 ~ .f C. H. Langford and M. L. Tobe, J . Chem. Soc., 1963, 506. Ionicstrength was not maintained constant with any supporting electrolyte; the solution was, however, rather dilute. S. C. Chan,J . Chem. SOC., 1964, 2716. K. S. Mok and C. K. Poon, Inovg. Chem.,1971,10,225. This work.At I = 0 . 1 ~ . S. C. Chan and M. L. Tobe,J. Chem. SOC., 1963, 514. S. C. Chan, Austral. J . Chem., 1967, 20, 595.7 S. C. Chan :mtl M. L. Tobe, J . Chem. Soc., 1962, 4531. M. E. Baldwin, S. C. Chan, and M. L. Tobe, J . Chem. SOC., 1961, 4637.u S. C. Chan, C. L. Chik, and B. Hui, J . Chew. SOC., 1967,d C. K. Poon and H. W. Tong, J . Chem. SOC. (A), 1971, 2151.S. ASperger and C. K. Ingold, J . Chem. Soc., 1956, 2862.f C. K. Lui and C. K. Poon, J.G.S. Dalton, 1972, 216.C. K. Poon and H. W. Tong, J.C.S. Dalton, 1973,1301.p S. C. Chan and M. L. Tobe, J. Chem. SOC., 1963, 5700.C. K. Ingold, R. S. Nyholm. and M. L. Tobe, J . Chem. SOC., 1956, 1691.TABLE 5Effect of leaving group on the acid and base hydrolysis of cobalt(II1) complexes of the penta-amine type,cis-CoLAX2+ (X = C1 and Br), at 25 "C Qk 2Br k2O1A k,Br/s-l klc1/s-l k lBr/ k lo' 1 mol-l s-l 1 mol-1 s-1 k2Br/ k6.5 x 1.7 x 3.8 1.4 a 2-5 x 5.6E%n32 2.2 x 10-6 5.1 x 4-3 4-8 x 10 7.0 6.9MeNH, 3-6 x 10-7 2.3 x 10-7 1-6 5.0 x 10 1-3 x l o e 3.9EtNH, 3.0 x 10-7 1.5 x 10-7 2.0 4.7 x 10 1.3 x 106 3.6PrnNH, 1-8 x 10-6 3.2 x 10-7 5.6 2.7 x lo2 1.3 x 101 21PrWH, 7.4 x 1 0 - 6 1.2 x 10-8 6-2 4.2 x lo2 5.2 x 1 O e 8.1CH,=CH.CH,.NH, 5.7 x 10-7 1.9 x 10-7 3.0 1-2 x 102 6.4 19CHsCCH ,-NK 3.2 x 10-7 1.3 x 10-7 2.5 5.7 x 10 3.3 17a Data from S.C. Chan, C . Y . Cheng, and F. Leh, J . Chern. SOC. (A), 1967, 1586, except as indicated; acid hydrolysis rate con-A. B. Lamb and J. W. Marden, J . Amer.dAt8 R. W. Hay and P. L. Cropp, J . Chem.stants were obtained by extrapolation; base hydrolysis was studied a t I = 0 . 1 ~ .Chem. Soc., 1911, 33, 1873.I = 1 . 0 ~ ; D. A. Buckingham, I. I. Olsen, and A. M. Sargeson, Inorg. Chem., 1968, 7, 174.Soc. (A), 1969, 42.F. J. Garrick, Trans. Faraday SOC., 1937, 33, 486; S. C. Chan, J . Chem. SOC. (A), 1967, 291.sensitive to the nature of the leaving group than a dis-sociative interchange mechanism of spontaneous aquationin which solvent participation occurs to a certain extentin the transition state. For bimolecular reactions, on theother hand, the breakage of the cobalt-halogen bond doesthe role played by the isothiocyanato-ligand in acidhydrolysis was rather extraordinary but the reasonbehind this remained a mystery.We thank The Society and the Committee O nOne of us (H. W. T.)not proceed far in the transition state and so the effect ofleaving group On a bimolecular reaction be muchHigher and Research Grants Of the ofHong Kong for financial support.received a Higher Degree Studentship from the Committee. smaller than that on a unimolecular reaction. Clearly,the above observation (Tables 4 and 5) denies a bi- [3/709 Received, 4th April, 1973