Redox Chemistry of [Fe2(CN)10]4£¾. Part 4.$ Reactionwith L-Cysteine%Floyd A. Beckford,a Deon Bennet,a Tara P. Dasgupta*a andGeoffrey StedmanbaDepartment of Chemistry, University of theWest Indies,Mona, Kingston 7, JamaicabChemistryDepartment, University ofWales Swansea, Singleton Park, Swansea SA2 8PP, UKL-Cysteine reduces [Fe2(CN)10]4£¾ to [Fe2(CN)10]6£¾ in a two stage process, a rapid reduction to [Fe2(CN)10]5£¾ followed by aslower second order reaction involving HSCH2CH(NH3+)CO2£¾ and a conjugate base.The binuclear FeIIIFeIII complex [Fe2(CN)10]4£¾ is a moderateoxidising agent with successive one electron reduction poten-tials1 (against NHE) of 0.55 and 0.36 V respectively.It is an inert, diamagnetic complex and reacts with manyreducing agents; with iodide reaction proceeds2 to theFeIIIFeII compound, but with ascorbic acid the additionalstage to form FeIIFeII has also been observed.3 When thereductant contains a nucleophilic sulfur centre as in S2O3 2£¾and (NH2)2CS then the binuclear complex may be cleavedand the iron(III) species [Fe(CN)5(Snuc)]n£¾ where n =4 forthiosulfate4 and n =2 for thiourea can be observed in theproducts.These species are readily recognised by a strongabsorption in the 550¡Ó650nm range, giving rise to a pro-nounced blue colour. The present paper describes a study ofthe reaction with L-cysteine, which is a powerful reducingagent (E 8=£¾ 0.34 V at pH 7) and also contains a nucleo-philic sulfur.Reaction was followed by monitoring the decrease inabsorbance at 560 nm, the wavelength of the visible maxi-mum for [Fe2(CN)10]4£¾.Preliminary work showed that therewere two steps, an initial fast reaction which was too rapidto follow under our conditions in which the [Fe2(CN)10]4£¾peak at 560nm disappeared and was replaced by the nearinfra-red peak of [Fe2(CN)10]5£¾ at 1200 nm. This was fol-lowed by a much slower process forming a pale yellow/green species with the spectrum5 of [Fe2(CN)10]6£¾.Thischange could readily be followed by stopped ow. The reac-tion was studied over the ranges 25¡Ó35 8C, pH 3.6¡Ó6.4.Individual runs for this second process gave good rst orderplots, yielding kobs/s£¾1 values that are directly proportionalto [L-cysteine] as shown in Table 1. The kobs values variedwith pH as shown in Table 2. At the higher acidities stu-died, pH 3.63¡Ó4.63, the change of kobs with pH is very slow,but at higher pH values the rate constant rises rapidly.Measurements made at 29.7 and 35 8C showed similar beha-viour.In view of the observations of Wilson et al.6 that traceamounts of copper catalyse the cysteine¡Ó[Fe(CN)6]3£¾ reac-tion we extended our original measurements to see if coppercatalysis is important here also.Wilson comments on thelack of reproducibility in earlier work on the ferricyanideoxidation of thiols, and inspection of the data in Table 1does show some scatter. We have analysed our reactants,buers and distilled water for copper by AAS and concludethat the copper concentration is less than 20 ppb (our esti-mated detection limit); this corresponds to 3.1510£¾7 moldm£¾3.In an attempt to complex copper (or other tracemetal ions) we added Na2[EDTAH2] over the concentrationrange 2¡Ó13010£¾7 mol dm£¾3, but did not observe any eecton kobs. Added copper (CuSO4) did produce a marked cata-lytic eect, plots of kobs vs. [Cu2+] being linear, with a welldened intercept on the [Cu2+]= 0 axis.Two other dier-ences from the work of Wilson et al. may be noted. For thereaction of [Fe(CN)6]3£¾ with cysteine pseudo second orderkinetics {w.r.t. [Fe(CN)6]3£¾} were observed whereas in ourcase all runs gave good rst order kinetics. Also for the[Fe(CN)6]3£¾ oxidations with cysteine, N-acetylcysteine and3-sulfanylpropionic acid Wilson found an apparent order in[thiol] that was greater than one.£¾dFeCN 3£¾ 6 =dtthiol ka kbthiol 1Our data in Table 1, though scattered, show no sign of anysystematic trend in kobs/[cysteine] with [cysteine].We con-clude that catalysis by adventitious copper has only a minoreect in our system, though with the addition of signicantamounts of copper there is undoubtedly some catalysis.Our oxidant, [Fe2(CN)10]4£¾, is a low spin, inert complex,and so we propose that the very fast initial reaction is dueto an outer sphere electron transfer to form [Fe2(CN)10]5£¾.This is followed by a slower outer sphere reaction to form[Fe2(CN)10]6£¾.The reduction potential for [Fe2(CN)10]4£¾is 0.55 V, considerably higher than the 0.36 V for[Fe2(CN)10]5£¾ thus providing much more driving force forthe rst reaction. While we conclude that there is little or noJ. Chem. Research (S),1998, 98¡Ó99$Table 1 Values of kobs for the reaction of L-cysteinewith[Fe2(CN)10]5£¾. [Fe2(CN)105£¾] = 1.710£¾4 mol dm£¾3; I = 1.0moldm£¾3(NaClO4); pH = 5.41 (acetate buffers); y = 30.0 8C103[Cysteine]/mol dm£¾3 kobs/s£¾1 kobs/[cysteine]10 3.21 3.2112 4.36 3.6314 4.59 3.2818 6.09 3.3822 7.18 3.0624 8.66 3.6126 9.46 3.64Table 2 Values of kobs for the reaction of L-cysteinewith[Fe2(CN)10]5£¾.[L-cysteine] = 0.020mol dm£¾3;[Fe2(CN)10]5£¾= 1.710£¾4 mol dm£¾3; I = 1.0mol dm£¾3 (NaClO4).y = 25 8CpH kobs/s£¾13.63 0.594.14 0.524.36 0.664.53 0.694.72 0.884.93 1.275.10 1.985.32 3.285.54 4.955.63 5.385.94 8.73$Part 3: see ref. 3.%This is a Short Paper as dened in the Instructions for Authors,Section 5.0 [see J.Chem. Research (S), 1998, Issue 1]; there is there-fore no corresponding material in J. Chem. Research (M).*To receive any correspondence.98 J. CHEM. RESEARCH (S), 1998copper catalysis for the second reaction we cannot excludethe possibility that it may contribute to the rst reaction,though the fact that the rst reaction was still fast in thepresence of added edta argues against it. In the absenceof kinetic data we cannot go further in discussing the rstreaction.For the second reaction the variation in kobs with pHmust surely be due to the acid/base equilibria of L-cysteine.It has three ionisable groups, carboxyl (CO2H), amino(NH3) and sulfhydryl (SH).In aqueous solution L-cysteinecan exist in ve dierent forms depending on the pH of thesolution [eqns. (2)¡Ó(6)]. The values for the respective dis-sociation constants7 are pK1=2.0; pK2=8.53; pK3=8.86;pK4=10.36; pK5=10.03.HSCH2CHNH3CO2H£¾£¾* )£¾£¾K1HSCH2CHNH3CO2£¾ H 2HSCH2CHNH3CO2£¾£¾£¾* )£¾£¾K2 £¾SCH2CHNH3CO2£¾ H 3HSCH2CHNH3CO2£¾£¾£¾* )£¾£¾K3HSCH2CHNH2CO2£¾ H 4£¾SCH2CHNH3CO2£¾£¾£¾* )£¾£¾K4 £¾SCH2CHNH2CO2£¾ H 5HSCH2CHNH2CO2£¾£¾£¾* )£¾£¾K5 £¾SCH2CHNH2CO2£¾ H 6Over our pH range it is clear from the pKa values thatthe main component of cysteine, more than 98%, isHSCH2CH(NH3)CO2£¾ with only minor contributions fromother species.The fact that kobs is almost constant frompH 3.63 to 4.36 at 25 8C suggests that the bimolecular rateconstant for HSCH2CH(NH3)CO2£¾ must be close to 0.59/0.02= 29.5 mol£¾1 dm3 s£¾1 and that undissociated cysteinehas negligible reactivity under our experimental conditions.The increase in kobs with pH must be due to contributionsfrom conjugate base species.Scheme 1 shows the proposed mechanism, and fromthis rate law (7) may be written, where A, B and Care HSCH2CH(NH3+)CO2£¾, £¾SCH2CH(NH3)CO2£¾ andHSCH2CH(NH2)CO2£¾ respectively.Rate k1A k2B k3CFe2CN105£¾ 7Now in our pH range [A] =[cys]T[H+]/([H+] + K2+K3),[B]= [cys]TK2/([H+] + K2+K3) and [C] =[cys]TK3/([H+] + K2+K3) where [cys]T is the total, stoichiometricconcentration of L-cysteine.On substituting into eqn. (7)and rearranging eqn. (8) was obtained.kobsH K2 K3=cysT k1H k2K2 k3K3 8When eqn. (8) is plotted using the data in Table 2 astraight line plot is obtained, and least squares analysisyields k1=27.020.9 mol£¾1 dm3 s£¾1 and 104(kK2+k3K3) =4.421.5 s£¾1. Treatment of the data by non-linear regressionanalysis yields 26.821.0 and 4.521.0 respectively. As theconcentrations of the tautomeric species B and C mustalways be in the constant ratio K2/K3 we cannot obtain sep-arate values for k2 and k3 except by some ad hoc postulatesuch as assuming that one of the tautomers is very muchmore reactive than the other.Thus if k2wk3 then k2 is1.5105 mol£¾1 dm3 s£¾1 at 25 8C. It should be noted thatalthough we have assigned k1 to HSCH2CH(NH3+)CO2£¾there must also be a small concentration of the tautomer£¾SCH2CH(NH3+)CO2H. From the fact that the pKa valuesfor CO2H and SH dissociation dier by some 6.5 log unitssuggests the concentration of this tautomer may be some106.5 times smaller than the main carbohydrate ionisedtautomer.However this would still give a bimolecular rateconstant well below the encounter limit. A similar argumentapplies to HSCH2(NH2)CO2H. Further analysis of the datais not justied. With other reductants that contain a nucleo-philic sulfur centre reaction with [Fe2(CN)10]4£¾ can give riseto coloured pentacyanoferrate(III) complexes with a sixthligand bound to iron through sulfur.We do not see thiswith L-cysteine, possibly because over our pH range the sulf-hydryl group is only slightly ionised, and reduction to[Fe2(CN)10]6£¾ is faster than nucleophilic substitution.ExperimentalMaterials.The complex, [Fe2(CN)10]4£¾, was prepared asdescribed previously.4 Solutions had max 560nm and concen-trations were calculated using = 1600 M£¾1 cm£¾1.L-cysteine hydro-chloride was supplied by Aldrich Chemical Co., USA and usedwithout further purication.Kinetic Studies.Reactions were followed by stopped-ow spec-trophotometry using a Hi-Tech Scientic SF-51 stopped-owattached to a Hi-Tech Scientic SU-40 spectrophotometer unit. Themachine was attached to a Haake GH constant temperature waterbath tted with a Haake D8 circulating pump. The pH was variedusing acetate and disodium hydrogen orthophosphate¡Ócitric acidbuers, the pH being measured with an Orion Research ExpandableIon Analyzer EA 920 tted with a Cole-Parmer combination elec-trode.Ionic strength was maintained at 1.0 mol dm£¾3 usingNaClO4. In all cases, the reaction was investigated under pseudorst order conditions with [cysteine]r[complex].Analysis for Copper.This was carried out using a Perkin Elmer2380 AAS instrument, calibrated with standard copper sulfate sol-ution. A good Beer Lambert Law plot was obtained up to 10 ppmof copper. All reagents including reactants, buers and distilledwater were checked and found to contain less than 20 ppb, ourdetection limit.This work was supported by the Board of PostgraduateStudies, University of the West Indies through scholarshipsto F.A.B. and D.B. We thank a referee for valuablecomments.Received, 15th September 1997; Accepted, 22nd October 1997Paper E/7/06692IReferences1 A. D. James, W. C. E. Higginson and R. S. Murray, J. Chem.Res., 1977, (S) 85; (M) 1084.2 G. Stedman, M. E. Pena and J. R. Leis, Transition Met. Chem,1992, 17, 123.3 F. A. Beckford, T. P. Dasgupta and G. Stedman, J. Chem. Soc.,Dalton Trans., 1995, 2561.4 T. P. Dasgupta, F. A. Beckford and G. Stedman, J. Chem. Soc.,Dalton Trans, 1993, 3605.5 G. Emschwiller and C. K. Jrgensen, Chem. Phys. Lett., 1979, 5,561.6 G. J. Bridgart, M. W. Fuller and I. R. Wilson, J. Chem. Soc.,Dalton Trans., 1973, 1274.7 R. E. Benesch and R. Benesch, J. Am. Chem. Soc., 1955, 77,5877.Fe2(CN)104¡V + cysteineFe2(CN)105¡V + AFe2(CN)105¡V + BFe2(CN)105¡V + C2A+¡E (or B+¡E or C+¡E)Fe2(CN)105¡V + cysteine+¡EFe2(CN)106¡V + A+¡EFe2(CN)106¡V + B+¡EFe2(CN)106¡V + C+¡Ecystinefastk1k2k3Scheme1J. CHEM. RESEARCH (S), 1998 99