摘要:

1976 2103Reactivity of Co-ordinated Amino-acids. Part 22 The Role of anUncharged lntermed iate in the Carbon-I 4- labelled-ligand-exchangeStudies on the frans(fac)-Bis(N-methyliminodiacetato)chromate(iii)AnionBy Andrew A. T. Bailey, Subrata Dutta-Chaudhuri, Charmian J. O'Connor,' and Allan L. Odell, UreyRadiochemical Laboratory, Chemistry Department, University of Auckland, Private Bag, Auckland, NewZealandThe rate of exchange of rrans(fac)-bis(N-methyliminodiacetato)chromate(iii) with frse ligand which has beenlabelled with I4C has been studied a t 50 "C, and c,, = 5.7 x 10-5-l.59 x mol dm-3. and shown to bedependent on the hydrogen-ion concentration. A mechanism involving partial aquation of one of the ligandsfollowed by re-anation has been shown to account for essentially the whole of the observed ligand-exchange rateimplying that direct bimolecular exchange is negligible.The contributing rate parameters at 50 "C have beencalculated. Confirmation of the mechanism i s provided by an anation study of triaqua(N-methyliminodiacetato) -chromium(tll) perchlorate at 50 "C. An uncharged intermediate in the aquation-anation equilibrium has beenisolated and equilibrium constants involving the three species in solution are evaluated a t 25 'C.WE recently studied the carbonyl- and carboxyl-oxygenexchange of tram (fac) -bis (N-methyliminodiacet ato) -chromate(m), [Cr(mida)J- (1), with lSO-labelled solventL -I-or(1)water and postulated a mechanism involving acid-dependent and acid-independent paths, both of whichinvolve a one-ended dissociated species.This specieswas not isolated.Evidence for dechelated intermediates in the acid-catalysed aquation of cis-bis(iminodiacetato)chrom-ate(m), [Cr(ida),]-, and of (1) has been reported byWeyh and Hamm who separated the 1+ forms of theseintermediates on cation resin columns. The productsof aquation, namely [Cr(ida) (OH,),]+ and [Cr(mida)-(OH,),]+, are also charged and this makes difficult theeffective separation of these products.absorption spectrum of the protonated intermediate inthe case of aquation of [Cr(ida),]- was indeed found tobe very similar to that of the triaqua-product but withslightly higher absorption coefficients. The experimentalevidence for a monoprotonated intermediate in thecase of the methyl derivative, [Cr(mida)J-, is somewhatmore tentative but its structure was postulated byanalogy with the [Cr(ida),]- system as (2), i.e.it is theproduct of cleavage of both C r O and Cr-N bonds in (1).Further evidence in favour of a one-ended dissociatedintermediate was presented in our lso-exchange studieson (l), in which the equivalence of rates of exchange of all1 Part 1, S. Dutta-Chaudhuri, C. J. O'Connor, and A. L. Odell,J.C.S. Dalton, 1975, 1921.J . A. Weyh and R. E. Hamm, Inorg. Chem., 1968,7,2298.3 C. A. Bunton, J. H. Carter, D. R. Llewellyn, C. J. O'Connor,A. L. Odell, and S. Y . Yih, J . Chem. Soc., 1964, 4615. * D. R. Llewellyn, C. J. O'Connor, and A. L. Odell, J . Chem.Soc., 1964, 196.The reportedthe eight oxygen atoms require dechelation of the ligandmolecule.We formulated intermediates for acid-depend-ent and acid-independent paths and postulated a formu-lation in which the Cr-N bond remained intact.It is frequently found that a CrIII-0 chelate bond islabile whilst a CrIII-N bond is inert. Thus 180-exchangestudies between H,180 and the tris(oxalato)chromate(m)anion and between H,180 and oxalic acid showed com-parable results,, indicating that ' one-ended dissociation 'of a Cr-0 bond was rapid compared with the rate-deter-mining addition of water. By contrast, hexa-ammine-chromium(II1) does not exchange its nitrogen ligands with15NH, in solution$ and only very slowly with liquid15NH, at 20 0C.5 Penta-ammineaniono-complexes ofCrIII usually undergo initial aquation of the aniono-g r ~ u p , ~ ~ ~ but a study of a series of variously substitutedacetatopenta-amminechromium(m) complexes showedr""\ /", H,O- Cr- N-MeMe HO=C-H,CrL( 2 )+that NH, substitution by water was the primary stepwith progressive strengthening of the Cr-0 bond andweakening of the Cr-N bond cis to the carboxylate, con-sistent with hydrogen bonding between the carbonyloxygen and adjacent H atoms of an ammonia ligand.Similarly, a nitrate group co-ordinated to CrIT1 activatesammine ligands in the cis position toward a q u a t i ~ n .~T. W. Swaddle, LaV. F. Coleman, and J. P. Hunt, Inorg.Chem.. 1963, 2, 950.6 F. Basolo and R. G. Pearson, ' Mechanisms of lnorganicReactions,' 2nd edn., Wiley, New York, 1967.C. S.Garner and D. A. House, Transition Metal Chem., 1970,6, 59.8 E. Zinato, C. Furlani, G. Lanna, and P. Riccieri, Inorg.Chem., 1972, 11, 1746.S G. Gunstalla and T. W. Swaddle, J.C.S. Chem. Comm., 1973,612 104 J.C.S. Daltonreports on the [Cr(ida),]- system, the spectrum of theintermediate is significantly different from that of thetriaqua-product. This leads us to believe that the CrIII-N bond remains intact in this intermediate, and we favourstructure (4a). The absence of hydrogen-bonding oppor-tunities in (1) also leads to the prediction that (4b) is un-likely to be formed. Further protonation and hydrolysisof (4a) might then lead to the breaking of the CrIII-Nbond leading to formation of (2)., We established thatour intermediate is transient in nature (see Experimentalsection).It is also clear that it is formed quite rapidlyduring the aquation of (1) (see above) and in the anationof (3) (see spectroscopic evidence below) in significantamounts. Any (2) which was formed during our identi-fication experiment would have been held on the cation-exchange resin together with the product (3).*Kinetics of [I4C]-Labelled-ligand-exchange Studies of[Cr(mida),]-.-Preliminary studies on the rates of ligandexchange of (1) with [14C]H2mida showed these to be com-parable to rates of aquation of (1) and anation of (3). Itis clear that meaningful determinations of rates ofligand exchange must be made under aquation equilib-rium conditions so that concentrations of (1) and (3) willThus the requirement for activating cleavage of a CrIII-N bond seems to be bonding from a hydrogen on the Natom to an oxygen on the ligand cis to it.In [Cr(ida)J-each NH group has four cis-acetato-groups and thus astructure analogous to (2) might well be formed through( 3 )SCHEME 1a mechanism involving hydrogen bonding. In thepresent [Cr(mida),] - system, however, hydrogen bondingis not possible and the CrIII-N bond is likely to be lesslabile.established the presenceof an equilibrium between (1) and its aquation product[Cr(mida)(OH2)3]+ (3), as in Scheme 1. We will showthat this equilibrium is important in formulating theligand-exchange reaction.Our 180-exchange studiesRESULTS AND DISCUSSIONWhile reinvestigating the aquation of [Cr(mida)J- wehave developed a method of separating an intermediateother than (2).The method of separation, which uses acation resin, indicates that the intermediate is not posi-tively charged. The experiment was made at pH 2 andit is unlikely that any dechelated species would remainunprotonated under these acid conditions. Unchanged(1) was eluted from the column and precipitated as itsinsoluble potassium salt. We therefore conclude thatthe remaining eluate, which contained 7% of the totalchromium and was coloured, consists of chromium(II1)species of zero charge. Two possibilities exist, (4a) and(4b) a'OI("\0-c=oMeH( 4 a ) ( 4 b )Comparative visible spectra of the initial anion (l), theintermediate (a), and the triaqua-product (3) are inFigure 1 where it may be seen that, in contrast to earliern1 I I I I 1350 100 450 500 550 600X / n mVisible absorption spectra of (a) 0.0085 mol dm'3[Cr(mida) (Hmida)], (b) 0.0084 mol dm-3 [Cr(mida) (OH,),]+,and (c) 0.0083 mol dm-3 [Cr(mida),]-FIGURE 1remain steady.These conditions were achieved by allow-ing a solution containing (1) and 95% of the desiredamount of unlabelled free ligand to reach equilibriumand then adding the remaining 5% of the H,mida in l4C-labelled form.Under these conditions plots of -loglo(l - F ) , whereF is the fraction exchanged, against time were linear forat least three half-lives. A typical plot is in Figure 2.* A full study of the anation of (3) is proceeding and willin-clude a search for (2) using Sephadex columns.Species (2) differsfrom each of species (1) and (ia) by only one water molecule andone proton, and recently A. T. Thornton, K. Wieghardt, and A. G.Sykes, J.C.S. Dalton, 1976, 147, have achieved effective separa-tion of the chromium(II1) complexes, [Cr(OH,),(0,CC,H4C0,)]+and [Cr(OI12),(0,CC,H4C0,H)]2+, which also differ only by onewater molecule and one proton, by using Sephadex separationtechniques1976 2105TABLE 1Calculated and experimental rates of exchange of tra~(fac)-bis(N-methyliminodiacetato)chromate(11r) anion and[1*C]methyliminodiacetic acid 0lO7R/mol dm-3 s-1102[Cr(mida),-] b 103[Cr(mida) (OH,),+] 104c~+ CEgnida * 1O'CHmIda ' Obs. calc.'5.0 0.57 0.10 0.4335.0 0.5 2.40 0.10 9.68 1.13 1.855.0 0.9 4.80 0.10 9.50 1.80 1.845.0 2.2 7.59 0.10 9.37 2.91 3.155.0 4.0 12.3 0.10 8.89 3.75 4.585.0 6.0 15.5 0.10 8.73 5.02 5.305.0 7.4 20.4 0.10 8.38 4.93 5.635.0 13.0 37.1 0.10 7.47 7.77 7.785.0 16.0 47.8 0.10 7.00 9.08 9.125.0 24.0 85.1 0.10 5.70 12.0 11.95.0 29.0 115 0.10 5.00 15.0 14.55.0 33.0 159 0.10 4.18 16.4 16.55.0 1.5 5.89 0.05 4.80 1.55 1.325.0 1.5 g 5.89 0.25 18.8 4.92 5.155.0 1.5 5.89 0.30 28.0 7.85 7.685.0 1.5 5.89 0.40 37.2 9.86 10.25.0 2.2 7.94 0.15 13.7 4.02 4.452.5 5.89 0.10 1.087.5 5.89 0.10 3.105.0 47.8 0.10 32.15.0 i 47.8 0.10 93.9a At 50 "C unless stated otherwise; all concentrations are in mol dm-3.ance at 480 nm unless stated otherwise.centration of [Wlligand.tH+. A t 65 "C.At 80 "C.Initial added concentration. Calculated from absorb-e Initial added con-0 Obtained from interpolation of plot of [Cr(rnida)(OHJ,+] againstMeasured after attainment of equilibrium and final adjustment.f Calculated from K , of H,mida.Rates of exchange, R, were calculated using a modifiedMcKay lo equation (1) where CL and c1 represent the totalcoefficients given in Table 2) and of Hmida- (calculatedfrom pKl 2.15 for H2midal1 and using pH measuredafter attainment of equilibrium).TABLE 2R == [ 2 ~ 1 ~ ~ / ( 2 ~ 1 + cL)]0.693/t+ (1)concentration of free ligand and complex (1) respectivelyand t = time. Rates of exchange under a variety of Molar absorption coefficients of [Cr(mida),]-, [Cr(mida)-(Hmida)], and [Cr(mida) (OH,),]+ at wavelengths of thet l h12 18 24 30 36 12 48 54FIGURE 2 Plots showing (a) the rate of exchange of 5.0 x lo-,mol dm-3 [Cr(mida)J- with 0.10 mol dm-, [1C]H,mida (initialconcentrations) and (b) the anation of 4.0 x 10- mol dmb3[Cr(mida)(OH,),]+ with 10.4 x 10-2 mol dm-3 H2mida at pH2.91 and 50 "Cconditions are detailed in Table 1.Also shown areequilibrium concentrations of (3) (calculated from ab-sorbances at 480 nm and using the molar absorptionisosbestic pointsr/dm3 mol-1 cm-1hlnm [Cr(mida)k)- [Cr(mida) (Hmida)] [Cr(mida)(OH,),]+480 39.0 39.5 15.4560 13.6 59.0 57.0A plot of R at fixed initial concentrations (c, = 5.0 xmol dm-3, cH,mi& = 0.10 mol dm-3) against CH+showed a non-linear dependence, and curved downwardsat high acidity in a manner analogous to that of thedependence of the equilibrium concentration of (3) oncH+.Plots of kobs. = R/~3~~mida against CH+ (a) and againstl / c ~ + (b) are shown in Figure 3. These results are consis-tent with a rate law of the form (2) which is identical toR = [(ko'/cH+) + b, + k2'CH+]C3CHmida (2)the rate law one would postulate for anation of complex(3) if this latter reaction were to occur through thepresence of H,mida, Hmida-, and mida2- whose respec-tive concentrations are dependent on acidity. Thus theanation rate may be represented as (3) which reduces toR(anation) = (kgcmida + KlCHrnida + k2CH,rnida)C3 (3)(2) if k,' = koKz and k2' = k,/K1 where Kl and K, arethe first and second acid-dissociation constants l1 of10 H. A.C. McKay, Nature, 1938, 142, 997.11 N. E. Ockerbloom and A. E. Martell, J . Amer. Cham. Soc.,1956, 78, 2672 106 J.C.S. DaltonH,mida. Values for k, were obtained from the inter-cepts of curves (a) and (b) (Figure 3) using a least-squaresprocedure and gave k, = 4.67 x lo4 and 4.00 x lo4dm3 mol-l s-l respectively (weighted mean = 4.17 dm3mol-1 s-l). From the gradients of the asymptotes, KO' =8.2 x 10-7 s-l and k,' = 4.43 x lo-, dm6 mol-2 s-l.Substituting these values and the equilibrium concen-trations of complex (3) and Hmida- into equation (2)gives us the calculated values of R quoted in the lastcolumn of Table 1. The agreement with the observedrates, while not excellent, supports the proposed kineticthe reaction shown in Scheme 2 we are measuring thetotal rate of formation of complex (1) whether it occursthrough (4a) or directly from (3).A typical plot oflogl,(At - A,) against t, where A and t are the absorb-ance and time respectively, is given in Figure 2.This kinetic run was made under identical conditionsto the exchange run shown in the same Figure. Forca. 1 half-life of reaction the plot is curved and thenremains linear for the next 2-3 half-lives. From thelinear portion the rates for formation were calculated andare presented in Table 3. They agree well with the rate10Lc,+/ mol dm-3concentration, and (b) [H+]-1FIGURE 3 Dependence of rate of [l4CJIigand exchange of [Cr(mida)J- divided by [Cr(mida)(OH,),+] x CHmida on (a) hydrogen-ionform and we may infer that the exchange reaction pro-ceeds through anation of the triaqua-species.This wasconfirmed by a limited study of the anation of (3) underidentical experimental conditions to those of the exchangereactions.Anation Rate Studies of [Cr(mida) (OH,),] +.-Solutionsfor these rate studies were made up as follows. The tri-aqua-complex (3) was present in a concentration equal tothat found at aquation equilibrium as detailed in Table 1.The H,mida concentration was made equal to the sum ofthat used in studies of the ligand-exchange rate and theconcentration of the acid liberated when the aquationproceeds to the same equilibrium position, i.e. equal tocH,,ia,(initial) + c3. After adjusting the pH, optical-density changes at 50 "C were followed at 560 nm.At(talSCHEME 2this wavelength (3) and any intermediate (4) whichmight be formed during the anation reaction have thesame molar absorption coefficients (Table 2). Thus inat which [14C]ligand appears in complex (1). Other dataare given in Table 3, and this trend appears to be general-ly true. The initial curvature we observe indicates thatthe path involving KIr is dominant, and that the forma-tion of a steady-state concentration of (4) controls theTABLE 3Comparison of rate constants of anation of [Cr(mida) (OH,),]+by H,mida at 50 "C with rate constants of exchange of[Cr(mida),]- with [14C]H,mida (corrected to first-ordervalues) under identical experimental conditions103[Cr(mida) (OH,),+] / 2.2 4.0 6.0 7.4mol dm-3102cH,m~~s/mo1 dm-3 10.22 10.40 10.60 10.74103cH+/m~l dm-3 0.76 1.23 1.55 2.04105k(anation) /s-l 12.8 9.38 8.37 6.66105k(exchange) * [Cr(mida)- 13.6 9.38 8.37 6.66* From Table 1.(OHJ3+]-l/s-lrate, Preliminary investigations at 480 nm show that K Iis very comparable to kII, and a full investigation of theeffects of concentration of complex, ligand, and hydrogenions on the anation reaction is in hand.This presentstudy has shown that in contrast to the [14C]ligand ex-change of [Cr(0,C,)3]3-, which occurred primarily by abimolecular mechanism with only a small contributio1976from a parallel aquation-re-anation path,12 in the presentsystem the whole of the ligand-exchange rate may beascribed to a process of aquation and re-anation of thecomplex in a reversible reaction involving the formationof an uncharged intermediate, and we may infer that anycontribution to [14C]ligand exchange by direct nucleo-philic attack on either [Cr(mida),]- or its aquation pro-duct [Cr(mida) (OH,),]+ is negligible.This difference provides yet another interesting illus-tration of the effect on reactivity of chromium(II1) com-plexes when the chelating ligands are changed fromdianiono-derivatives to amino-acid aniono-derivatives.We have already shown that introduction of the more2107(Charges on the species involved have been omitted inthe equations which follow.)At the two isosbestic points of 480 and 560 nm andusing the molar absorption coefficients given in Table 2,we have the relations (4)-(6).Using equations (4)-(6)15.4 cm + 39.5 c ~ ( H L ) + 39.0 c=, = A,,/d57.0 c n + 59.0 C ~ ( H L ) + 13.6 CHL, = A,,/d(4)(5)CML + CML(HL) + CML, = 0.002 (6)we calculated the concentrations of each of the threespecies containing M present at equilibrium and therebyTABLE 4Distribution of [Cr(mida)J-, [ML,], [Cr(mida) (Hmida)], [ML(HL)], [Cr(mida) (OH,),]+, [ML], H,mida (H,L), and Hmida-(HL) a and some equilibrium constants in a solution of total complex concentration = 2 x lo-, drn-,, pH = 3.05, andat 25 "C1 0 3 ~ 6 QH~O'C(H,L + HL) 1 O 2 c ~ ~ 104cML 1 04cnrt(HL) 1 0 4 ~ ~ ~ mol drr3 dm3 mol-'8 7.12 6.83 2.40 10.7 4.46 4.9314 12.5 4.72 2.92 12.4 4.24 4.9620 17.8 3.46 3.26 13.3 4.08 5.2925 22.3 2.62 3.85 13.5 3.50 6.60a All concentrations in mol dm-3.K' = C ~ ~ C H / C I L ( H L ) . PH = CML(HL)/(CMLCHL).robust chromium( 111)-nitrogen link allowed the lSOexchange of (1) with solvent water to take place by anacid-independent path, and this contrasted markedlywith the results observed for the chromium(II1) oxalato-systems where l*O exchange with water requires acid.In view of the importance of the intermediate (4) andof the aquation-re-anation equilibrium in the ligand-exchange process, we attempted to estimate the magni-tude of the constants involved in defining the precisepositions of the individual equilibria in Scheme 2. Dueto the nature and magnitude of the spectral changes in-volved, such a procedure can at best be only semi-quantitative.Spectrophotometric Determination of Equilibrium Con-stants involved in Formation of [Cr(mida),]+.-We con-cluded from spectral evidence of products that a solutionof [Cr(mida),]-, [ML,], on aquation reaches an equilib-rium with the dechelated intermediate [Cr(mida)-(Hmida)], [ML(HL)], and [Cr(mida)(OH,),]+, [ML], andhave shown from kinetic evidence that the anation of[ML] to [ML,] goes through the intermediate formationof [ML(HL)].Table 2 shows that an isosbestic pointexists at 480 nm between [ML(HL)] and [MLJ and at560 nm between [ML(HL)] and [ML]. The values ofc at these wavelengths were calculated from Figure 1.We analysed spectrophotometrically solutions of [ML](initial concentration = 2 x lod3 rnol dm-3) which wereallowed to come to equilibrium with H,mida (H,L)(0.084.25 mol dm-3) at 25 "C and pH 3.05.At thispH any free ligand would exist as ca. 90% Hmida- (HL)and 10% H,L. Since K, for H,L is 10-1O mol dmw3,l1the contribution from mida2- (L) would be small.l2 D. R. Llewellyn, C. J. O'Connor, A. L. Odell, and R. W.Olliff, J . Chem. SOC., 1964, 4627.calculated the formation constant, P H , and equilibriumconstant, K', for reactions (7) and (8). Results are givenin Table 4.BH[ML] + HL - [ML(HL)J (7)[ML(HL)] [ML,] + H (8)K'Measurements were made over a limited concentrationrange and the lowest concentration limit of added ligandwas imposed by the necessity for pseudo-first-orderconditions, i.e. the formation of [ML(HL)] or [ML,] from[ML] did not affect the total concentration of free ligandand hence alter the pH during establishment of equilib-rium.The calculations were made from small changesin small absorbances; thus at 480 and 560 nm the totalchange in absorbance in the equilibrium solutions we usedwas 0.062-0.072 and 0.068-0.056 units respectively.Although the values of K' and PH are only approxi-mate, they do indicate relative orders of magnitude forthese constants. A full anation study, which is in hand,will help to clarify the kinetic importance of these paths.Eflect of Temperature on the Ligawd-exchangeStudies.-Values of R observed at 50,65, and 80 "C for theligand exchange of c1 = 5 X mol dm-3 with CH,mida =0.10 mol dm-, at cH+ = 47.8 x 10" mol dm-3 are given inTable 1.The constant K, of H,mida is virtually inde-pendent of temperature in the range 0 4 0 O C . I 1 Wetherefore measured the pH at 50 "C and assumed thatvariation in these values with temperature and also asmall variation in the equilibrium position will notsignificantly affect our calculations of activation para-meters using the Eyring equation.13 This led to AH$ =l3 S. Glasstone, K. J. Laidler, and H. Eyring, ' The Theory ofRate Processes,' McGraw-Hill, New York, 19412 108 J.C.S. Dalton84.0 kJ mol-l and A S = -16.3 J K-l mol-l. These arecomposite values embodying contributions from rateconstants k,,', k,, and k2', but the value of AHt differsconsiderably from the approximate value of 53 kJmol-l evaluated from the [l*C]ligand exchange l2 of thetris(oxa1ato) chromate (111) anion.EXPERIMENTALMaterials.-Sodium trans (fac) -bis (N-methyliminodiaceta-to)chromate(m), Na[Cr(mida),], triaqua(N-methylimino-diacetato)chromium(m) perchlorate, [Cr(mida)(OH,),]-[ClO,], and N-methyliminodiacetic acid, H,mida, were pre-pared and analysed as previously described.' [14C]H,midawas prepared according to the literature l4 but using[14C]methylamine. The product was diluted with H,midabefore use so that its silver salt gave a specific activity ofca.360 disintegrations min-1 mg-l as measured in a Geiger-Miiller counter. Other reagents were AnalaR or werepurified before use.Apparatus.-pH Measurements were made on a Radio-meter (Copenhagen) 4 pH meter. The pH of anation reac-tions was a controlled by a Radiometer Titrator TTT2 (Copen-hagen).Stability-constant calculations were made fromspectra obtained on a Shimadzu QV 50 spectrophotometer.The extent of aquation in the exchange studies and calcula-tion of the equilibrium constants were obtained from visiblespectra recorded on a Cary 14 spectrophotometer. Anationstudies were made on a Varian Techtron 635 spectrophoto-meter using cells thermostatted by use of a Grant Thermo-circulator. Exchange studies were made in stoppered flasksin a Grant water-bath at 50 0.1 "C or in sealed ampoulesin oil-baths at 65 and 80 "C. The temperature was main-tained a t jO.l "C by using a Gallenkamp contact thermo-meter, Klaxon stirrer, and heating element. Carbon- 14specific activities were measured in Geiger-Muller or liquid-scintillation counters.Isolation of an Uncharged Intermediate in the Acid-catalysed Aquation of [Cr(mida),]-.-A solution of [Cr-(mida),]- (5 x lo-, mol dm-3 at pH 2.0) was held a t ca.90 "C for 1 h, long enough for attainment of hydrolyticequilibrium under the conditions used.The solution wasthen cooled and passed down a column of Dowex 50-WX8resin in the sodium form, followed by an equal volume ofcold water, and the eluate was collected and stored at 0 "C.This procedure removed the triaqua-product quantitativelywhile the starting material was removed from the eluate byprecipitating i t as its potassium salt (which has a low solu-bility) by adding excess of solid potassium chloride. Thesupernatant liquid was coloured and contained ca.7% ofthe initial amount of chromium present. Examination ofits absorption spectrum showed peaks substantially shiftedfrom those of the triaqua-product as shown in Table 5(along with data from ref. 2 for comparison).TABLE 5Details of wavelengths (A/nm) and absorption coefficients(E/dm3 mol-1 cm-1) for spectra of various chromium(m)complexeshnax.(1) Emax.( 1) L i n . Emin. Xrnax.(2) Emax.(2)trans-[Cr(mida),]- 363 28 405 11.6 493 42Intermediate 403 60.2 460 31.3 637 68.8[Cr(mida)(OH,),]+ 393 48.4 462 11.0 555 59.4cis-[Cr(ida),]- * 391 79 442 26 519 76Intermediate * 394 64 461 15 543 76[Cr(ida)(OH,),]+ * 393 50 460 13 564 60* From ref. 2 .After precipitating excess of potassium ions with sodiumperchlorate, the solution was allowed to stand overnight a troom temperature when optical densities had decreasedmarkedly, establishing the transient nature of this chromiumspecies.Determination of the Formation Constant of [Cr(mida) ,I-.-The ligand H,mida was dissolved in a solution of [Cr(mida)-(OH,),]+ and the mixture was titrated pH-metrically to pH3.05 before dilution to 50 cm3.The initial concentrationof complex was 2 x 10-3 mol dm-3, and that of H,mida wasin the range 0.08-0.25 mol dm-3. The solutions werethermostatted a t 25 "C for 100 h before spectrophotometricanalysis in the range 350-560 nm. During this time thepH remained constant.Carbon- 14 Exchange Studies.-Solutions of Na[Cr(mida) ,]and of H,mida (containing 95% of the total required con-centration) were mixed, diluted to nearly total volume, andthe pH was adjusted to approximately the required value.After establishment of the equilibrium between [Cr(mida),]-and [Cr(mida)(OH,),]+ at 50 "C (this was monitored untilthe absorption spectrum remained constant), the remaining5% of [14C]H,mida was added and a final adjustment wasmade to the pH by addition of not more than 0.4 cm3 of 0.3mol dm-3 HC1 or Na[OH] which had been equilibrated a t50 "C.The pH was monitored throughout each run and wasfound to be invariant. Two aliquot samples were withdrawnimmediately, one for determining the specific activity ofthe complex a t zero time and the other for evaluating spec-trophotometrically the extent of aquation. Samples (5 cm3)were removed a t time intervals, and the complex was separ-ated from the free ligand by precipitation as insolubleK[Cr(mida),] by addition of K[N03].The precipitate waswashed and was not contaminated by either the soluble[Cr(mida) (OH,),]+ cation or free ligand, even in the presenceof a large excess of free ligand (Found: C, 31.5; H, 3.7.Calc. for C,,H,,CrKN,O,: C, 31.5; H, 3.65%).The effective specific activity was too low to use a Geiger-Muller counter, and was therefore measured using liquid-scintillation techniques. The powdered sample (20 &- 2 mg)was suspended in a gel of Aerosil Cab-O-sil (Hoescht,Germany) in a counting bottle and the contents were uni-formly mixed. To this was added scintillation fluid [20cm3, made by dissolving 2,5-diphenyloxazole (6 g, Koch-Light) and 1,4-bis(phenyloxazol-2-yl)benzene (0.1 g, Koch-Light) per litre of scintillation-grade toluene (Baker)] andthe mixture was uniformly mixed.In runs containing ahigh acid concentration and therefore a large degree ofaquation i t was not possible to analyse 20 mg of sample. Inthese cases the specific activity was corrected from a colour-quench calibration curve of disintegrations min-l mg-lagainst the weight of sample (mg).The specific activity of Ag,[mida] (ca. 68 disintegrationsmin-1 mg-1) was determined similarly and infinite timevalues were generally calculated from equation (9). Thedisintegrations min-l mg-l at t, =weight of [W]H,midatotal weight of complexed and free midax specific activity of [14C]Ag,[mida] x 1.81correction factor of 1.81 was necessary because the specificactivity of K[Cr([14C]mida),] was 1.81 times higher thanthat of Ag,[mida] obtained from the same sample after de-composition of the complex and precipitation of the ligand(9)l4 G.J. Bershet, Org. Synth., 1938, 18, 561976 2109as the silver salt. Values calculated thus and thosemeasured on a sample which had been allowed to react for10 half-lives agreed within < 1%. All counting sampleswere prepared in the absence of diffused light and countedto 1% standard deviation.Carbon-14 exchange studies were followed in the rangepH 4.25-1.80 at 50 "C. At higher pH the rates were veryslow and at lower values almost complete aquation occurred.Half-times of reaction, ti, calculated from McKay plots lo of-loglo(l-F) against t were used to calculate the rates ofexchange R from equation (lo), where a and b are the con-centrations of free and complexed midap.The error in theestimate of R is f 2%. Limited data were also obtained a t65 and 80 "C.Kinetics of Anation at 50 "C.-The required volume of[Cr(mida) (OH,)3][C10,] solution was added to a solution con-taining a weighted amount of H,mida. All the solutionsand equipment were allowed a t least 45 min to attain tem-perature equilibrium before mixing. The pH was adjustedto the required value by addition of 0.5 mol dm-3 Na[OH]and maintained a t this pH k0.005 using a Radiometer pH-stat-titrator, based on the pH meter 22. In order to avoidprecipitation of K[Cr(mida),] as the reaction proceededbecause of continuous contact of the calomel electrode withthe reaction solution, a modified Laitinen salt bridge,l5 con-taining 1 mol dm-3 NaCClO,] and NaCl sealed within aU tube by two small (diameter O . A O . 6 cm), porosity 4, sin-tered-glass discs, was used to connect the saturated calomelelectrode to the glass electrode. Diffusion of Na[OH]from the injecting tube due to continuous immersion waseliminated by use of a plunging device which only operatedthe injector when the pH fell below the set value. Dilutioneffects caused by additionof Na[OH] solution were minimisedby using a relatively concentrated solution. Throughouta run no more than 0.5 cm3 Na[OH] per 50 cm3 of reactionsolution was added. Aliquot samples were withdrawn witha syringe and transferred to a dry equilibrated cell. Changesin the visible spectra were recorded as a function of time.The reference cell contained water. Pseudo-first-order rateconstants of anation k+ were calculated from a plot of log,,-(A, - A,) against t, where At and A , are absorbances a t560 nm a t times t and infinity respectively. ' Infinite ab-sorption measurements were made 12-24 h after com-mencement of the reaction depending on the hydrogen-ionconcentration. These readings were checked intermittentlyfor constancy. Each experimental run was duplicatedand the error in the estimate of k4 was f2%.We thank the Research Committee of the New ZealandUniversity Grants Committee for support, the University ofAuckland for the award of a postdoctoral fellowship (toS. D-C) , and Drs. R. W. Olliff and D. Shooter for helpful dis-cussions.[5/2510 Received, 22nd December, 19751l5 K . A. Laitinen, Ind. and Eng. Chem. Analyt. Edn., 1941,18,393

ISSN:1477-9226    

DOI:10.1039/DT9760002103

出版商:RSC    

年代:1976

数据来源: RSC