<%BANNER%>
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Experimental
 Results
 Discussion
 References
 Biographical sketch
 Back Matter
 Back Matter


UFIR












INTRODUCTION


Althou[lh the coordination of metal ions by sulfur in the

form of mercaptide, sulfide, or thioether functions have long

been known, only during the last decade have their interactions

been recognized as vital in biological systems. Metal-sulfur

coordi'r t.ion is -"o kc.i'nc to plpy key roles in the processes of

phoosynt ;iis, nitrogen fixation, oxygen metabolism, hydroxy-

lation of steroidal compounds and electron transport, Also

several compounds vhose specific biological function has not yet

baen elucidated are know-n to have metal-sulfur coordinat.ior.2

The most explicit demonstration of metal coordination by

su3fur has come from recent structural determinations by x-ray

crystallography. In clostridial rubredoxin, whose function is

unknown, the single iron atom is coordinated tetrahedrally by four

cysteinyl .crcaptide functions.3a Spectral studies indicate that

this coordination is retained in solution.3 In horse heart and

xbonito crytochrome c, the heme iron is coordinated in the out-of-

plane positions by imidazole nitrogen and methionyl thioether

functions. Thus the importance of both mercaptide and thioether

eoor~(ni ion is c''iit elyr cs'tb2ished.

L L. :ivo C;; 10. and physical investigations to date have

L i C-.~ ~. oaC t O. O;.; ,L2ers of thl clas Oi non-hene iioI proteins

(lTi;') !:: 'n as icr-. (cins viich are involved in yhotoyiihetin




2



and nitrogen-fixation processes. 15 The isolation of several

proteins of this type has formed the basis of investigation work.6

In general, the compounds of this class contain stoichiometrically

related non-heme iron, cysteine, and acid-labile sulfur, exhibit

physical parameters which are anomalous for iron and serve in

biological electron transport functions. Specific familial

characteristics of ferredoxins are (1) their relatively low

molecular weight (-12,000-30,000 g/mole), (2) the presence of acid-

labile sulfur (treatment with acid produces H2S) in an amount

approxi-terly-. equivalent to their iron content, (3) s-oi~io-

metric relationship of iron and cysteinyl ligands, (4) electronic

absorption and electron paramagnetic resonance which is anomalous

for iron complexes and (5) oxidation potentials (0.2 to 0.4 v at

pH 7) which are unprecedented for iron complexes. Chemical

subunits of this class are also incorporated, with other redox

functions, into more complex enzymes which are utilized in various

biological redox processes.

The presence of both iron and labile sulfur at the active

site of a typical two-iron protein was established by cpr measure-
2b
ments on isotopically substituted species. Iron and acid-labile

sulfur were removed from the protein to yield the inactive apo-

protein. Biological activity was restored by treatment with iron

salts in combination with 2-mercantoethanol and inorganic sulfide.

Subsit ti n of ocleri.u for minor 'nic sulfur also rccn(mrated a

subsatrLiaiiJv active ?ein.7 ', s plittino of the c)r signal

in die reuucid protein ~y indepenuie&ly substitLted appropriate

isotopes of iron and sclenim e st- bs hed theii mutual proximity








to the site of reduction. A similar proxil-ity for the cysteiryl

sulfur atoms is suggested by the epr behavior of protein produce-

by orgarlisns gro'm on an isotopically substituted source of

sulfur.

This brief summary is representative of recent research

which demonstrates an extensive and varied utilization of met"l.

ions coordinated by thiolate, thioether and "labile" sulfur atc s

in biological oxidation-reduction processes. The research to Le

described here represents an effort to examine the influence of

two of ::-- donor ".fuctions on the redoz behavior cf certain

metal ccr lexes. T'he complexes were chosen not for the exteit to

which they simulated the biological examples but rather for their

virtue of incorporating certain of the biological aspects into

systems rihose reactions stood the best chance of being both

thoroughly characterized and interpreted in relationship to prior

fundamental studies. This objective stands in contrast to the

alternative approach of investigating systems more directly related

to those found biologically. However, the iron(III)mercaptide

complexes suggested by the biological systems are complicated,

unstable and kinetically labile, leading to less than definitive

results. Simple complexes wVth iron-thioether coordination havw

proved elusive to synthesis and would likely present a low susccp-

tibility to thorough kinetic description.

r7he reaction! described hIerein serve as modil systems in

only tio. Inot Judir -iay scnu that they do incorp:'. i: the iio-

logical Lypt of Luonor functions as ligans Tie po..Luilit.y o

substantial. differences betutciu the reactivity pat.:-r1 i d- i .:.s i








here and tho'e of the biolo i:cal c:irples is openly anticipated ;

nevertheless. a Petter understanding, of how these ligands affect

oxidation-reductonn behi-ior should shed some lig t on the reasons

for their extensive biological utilization.

One approach to evaluating the effects of mercaptide ztid

Ihioether coordination involves the study of well-characterized

complex es whose reactors are susceptible to detailed mecl-anistic

investigation., further ionomeric, inert complczes of cobalt(!11)

or cho-or..i::(3 (I') containing" a sin .e coordinated 2p arscptide or

thin'-':" rction aporr: desire in the Initin.; s uri"s f

comparative roses L .ith earlier lith c i.s obj. i'es a

research project in these laboratories result( in the preparation

and isolation of LCo(en)y2 (OCCH2 For the purpose of

conp.ring the reactivities and mech.rnistic patterns an analogous

co.mp'ound with the sulfur replaced by a nore classical oxy;cen donor

atom was prepared, [Co(en)2(OOCCH20) l 4.9 For the pair of can-

plexer, the behavior toward chromiu~(lT) and [Ru(iM 3) ]2+ as

typical inner-and outer-sphere reductants, respectively, as well

as applicable substituti.on7 behavior of the reaction products

was cxa zined.' This .arked the first significant developrment

in evaluatin- the infle:ice of a cce dinated nercaptide in oxida-

tion-reduction reactions.

In order to consider these r 'cults and those described

h"r~. i-t is n-o r reo'ni. 7 t two -'ncr~al cate'rories

";:-e'. t ..-.i: bl : or 0 '' .on-rc ,..on rac aions of

ts, ...... r s '' .: ** sc;- i;.hi ii, :i i r -act o rte
]i
%:. r,, ]: ? ".." .'. ..i..C :.; e ]iOi,: oS: 1'r-: r ~e to% n, i~ie








two reacting metal centers are joined in the activated complex

via a ligand cornon to the first coordination spheres of both

metal ions. An outer-sphere reaction is characterized by an

activated complex in which no sharing of lirands occur, that is,

the primary coordination spheres of the respective metal ions

remain intact with no bond cleavage or bond making required for

electron transfer.

For an inner-sphere reaction one of the reactants must be

sufficiently labile so that ligand substitution involving a ligand

from the second reactant can occur prior to electron transfer. The

bridging .igand(s) of the second complex is required to have

available electron pairs or orbitals of sufficient energy and

proper symmretry for forring a bridge between the two netal centers.

The bridging ligand brought in by the second reactant must be re-

ained long enough for the electron transfer to occur. By meeting

the above requirements an inner-sphere reaction is made possible

but not mandatory. An outer-sphere reaction may still provide a

path of lower energy.

A reaction is most conveniently assigned to the inner-sphere

category if the bridging ligand can be detected in the coordination

sphere of the product of the labile reactant at higher than equil-

ibrium levels. This requires that it be retained due to an inert-

ness of this product to substitution relative to the rate of the

redox reaction. Exenalifyinf this means of categorization is the

classical reaction t-btw n the substitution-inert complex, [Co(:-

H )5(01)]+', and substitution-labile [crC!20). in acid solution
+ ubtitution-labile Co -20)2+
to yield substitution-lhbile [Co(;tc)I and substitution-







inert [Cr(H20)5(Cl)]2+.12 The capture of the chloride ion in

the first coordination sphere of the inert chromium(III) product

is definitive of an inner-sphere reaction. When the same

reaction is carried out in the presence of radioactive free

chloride ion, no radioactivity is found in ite chromium(III)

product, thus elir.inating its incorporation prior or subsequent to

electron transfer which identifies the coordinated chloride in the

product -s originating with the cobalt(III) complex.12 Other

donor functions ihich have been identified as bridging ligands for

cobalt(III)-chronium(II) reactions are the halides, carboxlates,

azides, thiocyan~tes, phosphate, sulfate and hydroxide.2'13 In

contrast, an outer-sphere reaction is decisively dictated if one

of the reactants is substitution-inert relative to the rate of

electron transfer and contains in its first coordination sphere

no ligand capable of bridge formation. A well-characterized

example is the outer-sphere reluctant, [Ru(i3)612+.14

The invest tion of the mercaptide and alkoxide complexes

previously describe d established, through characterization of the

chroniur(III) pr oducts, that the reactions of the species with

chroriu,(II) occur via inncr-sphere paths.l0 The reduction of

[Co(en)2(OOCCH20)1] was concluded to occur via a bridging reaction

utilizing the a Woxide or from kinetic comparisons. It is

directly comparOble, therfo e, to the reduction of [Co(cn)2(OOC-

CH2S)I+ which r-oc: ded -i a mercaptide bridged path as esta-

blished by product charact rization. The results clearly esta-

blished a reactivity tovrd chrominun(II) for the sulfur complex

which is '103 tiire3 Teatet than for its oxygen analogue.








While the previous research established that the coordinated

mercatide exerted a substantial increase in reactivity for oxida-

tion-rcedction, several important questions remained unanswered.

The various factors which night possibly contribute to thi: enhanced

reactivity rernaind unevaluated though recognized. In particu-

lar, the relative influence of steric vs electronic factors could

not. be determined. It seemed desirable to evaluate these influ-

ences in cormplexes as similar as possible in order to iinirmize

any effects ari.lnr.g from the contribution of the standard free

er.c---. c-.-" t t- reactiv'ties.15 T17-, if a rrcr.ter ster-c

acci. 1i; ili y of ti larger sulfur atom were mainly;, responsible for

the observed reactivity pattern in otherwise comparable complexes,

it was felt substitution of a methyl group fcr one of the adjacent

methylene protons should decrease the rate reduction cf the sulfur

conplex less than that of the analogous oxygen complex. Further,

if the anticipated rate decrease is observed, the possibility

would arise of determining the enthalpies and entropies of activa-

tion which were not accessible for the mcrcaptoacetato complex.

These should shed further light on the steric and electronic con-

tributions to the reactivity differences. Thus, the complexes

[Co(en)2 (OOCC (:. ,)3)](C104) and [Co(en)2(OOCCH(CE )0)](C104) were

prepared in order to further define the influence of coordinated

mercaptide.

It ws ltS ho~, that the ahove modification in the alkoxide

lip.,:J : I a rore dcJi native ch.a'acteriz Lion of the

ciu.>,,ii) p~. ;ct as diuriving fro-i alkoxide bridging ltan was

po it i: iou :" investir::.Lc' analo ic. ,'hen this vas








found not to be the case, a complex having a chelate ligand which

contained a carboxylate function as the only possible bridging

group, [Co(en)2(00CCH2 2t +,)] was investigated in order to iso-

late any unique effects arising from chelation relative to those

previously established for monodentate carboxylate coordination.15

Comparison utilizing results with this complex were expected to

more rigorously define, and possibly exclude, the participation

of carboxylate bridging for the alkoxide-containing complexes.

In spite of its demonstrated biological importance the redox

influence of coordinated thioether functions remains poorly under-

stood, The first kinetic study directly involving this mode of

coordination in simple systems suggests that the effect of thio-

ether donor functions as non-bridging ligands in an inner-sphere

reaction is to enh1anc reactivity.7 Earlier rate studies cf com-

plexes such as [Co('> ) )(OOCC:2SCH2Cg H)] suggested that a pendant

thioether enhanced electron transfer rates above that observed

with the coordinated carboxylate alone, presumably via chelation

of the reLiuctant.18 In anticipation of this possibility and in

collaboration with concurrent research19 the complex [Co(en)2,

(O0CCH2SCu 3)]2+ was prepared and studied. .Whatever the mechan-

istic pat:w ay taken, it was expected that the similarity of this

complex to the previously studied mercaptide precursor should

provide a better understanding of the influence of coordination

by a thioether donor.

Thus, the objectives of this thesis include (1) evaluation

of the steric and electronic factors responsible for the increased

reactivity of mercaptide complexes relative to their oxygen ana-




9



logues, (2) further comparison of these influences within the

respective alkoxide and mercaptide classes of complexes, (3)

definition of the role played by carboxylate incorporated in

chelate system as opposed to the monodentate carboxylate function

and (4) an investigation into the accessible aspects of thioether

coordination with the additional complex nade available by this

research. Further, the behavior of the chrmriun(III) complexes

uniquely produced by reactions originally investigated for their

relevaz ce to oxidation-reduction chemistry should contribute to

a bt'l' r-e.''- inr of tbhe ubstituticnal characteristics of

LthQe; ii ",iands.






















R-Fcents --Common chemicals were of reagent grade and were

used 'ihout further purification unless otherwise specified.

-t .-Dis.-.ll]ed water used in kinetic exncri,'rent and in

p-eparation of v i.ious stock reagent solutions was obtained by the

distillation of deionized water from alkaline permanganate solu-

tion .-sing an all-geass distilling assembly and stored in poly-

stoppered, glass bottles.

!jitro-en.--For deaerating solutions of air-sensitive materi-

als, line nitrogen was passed through two successive scrubbing

towers containing Cr(II) solutions which were prepared by reduc-

ing 0.1 r chromiu.(III) ion in 0.4 M perchloric acid in the towers

with a bhd of aral -s.ted zinc. To assure the nitrogen was satur-

ated with water, it was then passed through a tower containing

redistilled water.

Zinc a m.alenm.--Twenty mesh granular zinc (Fisher) was activa-

ted with 3 M1 HC1. After several rinsings with distilled water the

activr-' -' 'no v -.I cra,'It IinF- a solution of tetrachloro-

'-r,- 'T) in' i" 1 iP.1 fcr ten sccnd. After s.,veral wash-

ir..cL .. u^~it-l.- ...u-r Ai-~i t ciCn ~~was uri.ed under a sreaam of


EXPERIMENTAL








Lithi' rercrh:orate.--Reagent grade lithium perchlorate was

used throughout to maintain constant ionic strength, Purity was

checked by passing a prepared solution through an ion exchange

column in the acid form and titrating the collected solution with

0.1 M sodiv:; hydroxide to a phenophthalein end point.

Chroiiil)_ ioni.--Reagent grade chromium(III) perchlorate

(G. F. Smith) was used to prepare stock solutions ranging in con-

centration from 0.050 M to 0.25 LM. Aliquots of the stock solutions
-4
were diluted to concentrations ranging from approximately & x 10
1- -h
i" i 2 x T O in Pol"tions vary-in in acidity frem 1 :c 10 :

to 1 I. roIducticn to chronium(II) was accomplished using zinc

amalgam.

Chlororentz.anin3'obalt(ITI) chloride.--This compound was

kindly made by Mr. Peter F. Eisenhardt, who used a standard pro-

cedure for its preparation.20 The compound exhibited a molar

absorbtivity of 49.9 at 534 nm in excellent agreement ,rith the
21,22
values of 50.2 51.0 previously reported.2122

Mercantoacetatobis(ethylenedianine)cobalt(III) perchlorate.--

This compound was generously provided by Dr. Robert H. Lane, by

whom it was first prepared and characterized. The sample provided

exhibited the following; spectra] characterization: [\A(): 518(1521

282(11,700)1.

2.2 1 -TD it,. i -.r.-. -".:n acid. --2-mercaptooronionic acid

(l'drich C' -Acal Co., reagent Trade) was converted to the disul-

fi ific-tion of Uie ,:.ithod of Fredga and Bjorn- which

appVaZCt Lu ugc gef ral fr preparation of simple disulfides from the

corres'onrin mrrcapa def, 24 One milli::ole of the ncrcaptan







was slowly added to a solution of one nillimole iron(II) per-

chlorate dissolved in 200 mil of water. After addition the solution

was allowed to stir for one hour, several drops of concentrated

sulfuric acid ias added, and the reaction mixture extracted with

three 50 rm portions of diethyl ether. The combined ethereal

extracts were evaporated to dryness under a stream of nitrogen.

The resultant white solid exhibited a melting point of 113C

uncorrected (lit. 1140 5.00C)23 and was used without further

purific tion.



Fr 1 -.. ni:' C. -.a.



2- ..- ..i , ir 'c..:. ti'III'I : i .j:- .-_ --
[Co(ern) (OOCCI (, S)]C0

Sl" carsti., 3f tC is cc:':o'rund pCi-al-lcd -the- mthod usd in

preparation of the mercaptoacetatobis(ethylenediamine)cobalt(III)

complex previously reported.9 Ten grams (.028 mole) of Co(C104)2'

6H20 (G. F. Smith) was dissolved in 30 ml of H20 in a 100 ml

three-neck flask fitted with rubber septum stopper, nitrogen inlet

and outlet, a n magnetic stirrer. After deacration for thirty

minutes, 3,6 ml (.057 mole) of 98-100 per cent ethylencdiamine

(Baker) was aJdod by syringe. After another thirty rinutesideaer-

ation time, .. g of solid 2,2' -dithiodipropionic acid was added

by r2i-a rec,-)- j.nd replaci.ent of the nitrogen outlet tube.

within five minutLs tne solid had dissolved and the solution color

c'"'. 2-' a ': color. ACter thirty rhin-tcs under

n r:oe e i tion was tr-ansierred to a 150 ml beaker, and







evaporated with stirring on a hot plate at 100C under a gentle

stream of nitrogen until the volume was approximately 20 rml then

allowed to cool to ambient temperature. The resultant foamy brown

mass ias filtered and washed with 5 ml of hot water. Recrystalli-

zation was effected by dissolving the collected solid in 100 ml of

hot (90C) water, filtering, and cooling in an ice-acetone slush.

Light purple needles separated on cooling. The product was re-

cryjtallized twice more as described, washed with two 10 ml por-

tionr of absolute ethanol and one 10 ml portion absolute diothyl

ether, tnen dried in vacuo over CaSOc for twelve hours. Yield

2.5 ;-. Anal. Crlcd. for [Co(:-C, CH2 c.2 ) (OCO (CcT )S)CI10 :
2 222 3'JtV
C, 21.97; H, 5.27; N, 14.64; s, 8.36; Co, 15.40. Found: C, 21.75;

H, 5.?1; 14.61; S, 8.32; Co, 15.15.




Eighteen grams (.05 mole) of Co(C104)2 6H20 (G. F. Smith)

was dissolved in 125 ml of water in 250 ml Erlenmeyer flask fitted

with a two-hole rubber stopper with glass tubing of appropriate

lengths. Six grams (.10 mole) ethylenediamine (Baker) was added

witl stirring, then air ':as dr.wn through the solution for twelve

hours. To this solution was added a solution of 5.05 g (.05 mole)

70,^ aqueous lactic acid (Baker) and 2.00 g (.05 mole) NaOH pellets

(FSsher) in 20 Mi. water. The combined solutions were transferred

to a 2L0 ml tn w er and e'vaporatd undtr a gentle stream of nitro-

gc-n .iu stirring on a hot plate at 103O'C uniil the volume was

ap:-,',:':cely 310I rl (cc. lO i"'n). It mi: cooled to imbient

te r taLure and ,ne solJid preent were filtered ana washed with







10 nl water. The combined filtrate and washing was returned to

the hot plate and evaporated as before to a volume of 50 rrl,

then cooled to ambient temperature. The resulting solid was

filtered, washed with 5 ml of water, 15 ml of absolute ethanol,

and 15 ml of absolute diethyl ether and dried in vacuo for ten

hours over CaSO4. Yield = 4.75 g. Anal. Calcd, for [Co(NH2-

CH2CH 2 I)2(00CCH(CH 3)0)C104: C, 22.90; H, 5.52; N, 15.27;

Co, 16.11. Found: C, 22.86; H, 5.52; N, 15.14; Co, 15.82.

2-Methylttioacetatobis(ethylenediamine)cobalt(III) Diperchlorate--
[Co(en)2(OOCCH2SCH )](C104)2

To a suspension of 1.8 g of [Co(en)2(OOCCH2S)] in 300 ml

of 904 methanol-water mixture was added a large excess (25-fold)

of methyl iodide (Baker). The mixture was stirred in a closed,

round-bottom flask for twenty-four hours and for three nours was

stirred while open to the atmosphere. The light pink solid was

collected, washed with two 25 ml portions of absolute ethanol, two

25 ml portions of diethyl ether and dried in vacuo over CaSO4 for

twelve hours. The solid was recrystallized by dissolution in

minimum amount of hot (900C) water followed by an addition of

solid EaClO1 (~4 g) until precipitation began. After allowing it

to cool to ambient temperature, the bright red-orango solid was

collected and recrystallized again. Yield = 0.70 g. Anal. Calcd.

for [Co(lH2CH2CH2UNI )(OOCCH2SCH )](C104)2: C, 17.40; H, 4.38;

N, 11.60; S, 6.64. Found: C, 17.42; H, 4.26; N, 11.64; S, 6.75.








-*I!* (TI "1 -iT r;[ -r t. .1:f


This compound was prepared by the reaction of silver perchlo-

rate with glycinatobis( ethylenedianine)cobxlt(III )dichloride.16

Twenty grars of trans-[Co(en)2Ul2]Cl was suspended in 40 ml of

water. To this was added 8.1 g silver oxide (Matheson, Coleman,

and Bell) and the suspension was ground in a mortar and pestle

periodically for one lour. Silver chloride was removed by filtra-

tion and iwasned with 40 ml of hot (90C) watcr. To the coribined

fi ltrate and washings was added 6.9 p glycine (Fisher), and then

th, rix:t-'rc was evarorz.ted on a steam bath ndcr a ctrea~ of nitro-

gen to a thick syrup. After standing at ambient temperature, the

rea-.ultart ;olid was filtered, washed with cold water, rccrystallizud

from hot water, and dried in vacuo over CaS04 overnight.

A solution o01 u.o g (o.ul mo-le) o01 [o(en)2(0OCCGl2iiH2)]jC12

dissolved in 40 ml water was added slowly with stirring to a silver

perchlorate solution prepared by adding 2.36 g (0.0085 mole) silver

carbonate (Mallinkrodt) to 10 ml of 2 M perchlorate acid. Silver

chloride was removed by filtration and the filtrate was reduced on

a rotary evaporator at 45C to a thick syrup. After several days'

standing at ambient temperature the resultant solid mass was filtered

and air dried using aspirator suction for 2 days. The solid was

dried in vacuo over CaSO4 for tw;enty-four hours. Yield = 5.0 g.

nnl. cl. i. for [Co(NICo ..i )(.* C' )'(Cl1 ) ;: C, 15. 4;

'. LIt,: *1 15;.49; Co, 1).0' 4. C, J'jt; H, I4. ;2-; 15.52;








Chromnium(III) Complexes of 2-Merca.ptopropionic Acid. Lactic Acid
and Glycine

Several aquo-chromium(III) complexes with these ligands were

generated in solution by the reduction of the cobalt(III) complexes

with chroamium(II). Some of the complexes so produced underwent

further substitutional changes, the products of which were separ-

ated according to charge type by ion exchange chromatography and

further characterized by their ultraviolet and visible spectrum

and by chromium analyses. Synthesis and isolation of the complexes

was not a primary objective; therefore, discussion of them is

deferred to the section on Results.



Analyses



Chromium(III).--Determination of chromium(III) was accomplished

spectrally by alkaline peroxide oxidation to chromate(VI) ion which

was monitored at 373 nm (E = 4,815 15).25 To an aliquot of the

chromium(III) complex, 10 .1 distilled water, 10 ml 0.20 M aOHil

and 3 mll of 30& hydrogen peroxide were added. The oxidation

usually was complete overnight. Excess peroxide was deconposed

by heating the solution at 600C using a coiled platinum wire as

catalyst. When cool, the solution was diluted to 100 ml and

absorbance at 373 no was observed. Duplicate runs usually were

reproducible to within 1%.

Chromium(II).--Periodically aliquots of chrorium(II) solu-

tions were reacted in an inert atmosphere with a solution contain-

ing a known excess of chloropentaa&ninecobalt(IlI) in 0.1 1 HC104.

The excess was determined by analyzing spectrally the resultant








solution at 534 irn, after subtracting for absorbances due to

presence of hexaquocobalt(II) ion (C,4 = 3.1)26 and chloropenta-
27
aquochromium(III) ion (534 = 5.5).

Cobalt.--The method of Moss and Mellon using 2-2',2"-ter-
28
pyridine as a completing agent was used.2 A sample of complex con-

taining approximately 4 rag cobalt was destroyed using 15 ml liquid

fire reagent (7 parts 70% HC104 3 parts 70% 11NO ). The reaction

mixture was evaporated just to dryness on a hot plate, with the

residue then dissolved in 20 ml distilled water. Addition of 5 ml

of 20% axnonium acetate solution brought the pH to about 6 and the

solution was then diluted to 100 nl. A 25.0 ml aliquot of the

cobalt(II) solution was deaerated using a 50.0 volumetric flask

fitted with septum stopper and hypodermic needles for nitrogen in-

let and outlet. Ten riilliliters of 0.2 terpyridyl solution '.wac

injected and volume brought to 50.0 ml by syringe addition of

deaerated distilled water. A sample was then transferred to a

deacrated septum-stoppered 2 cm cell using syringe techniques and

the concentration of the cobalt(II) terpyridyl complex determined

spectrally at 505 nm. A molar absorbtivity of 1,386 at 505 nm
29
determined with known cobalt solutions29 was used rather than the

reported value of 1,360.28

Elemental analyses.-Analyses for carbon, hydrogen, nitrogen

and sulfur were performed by Galbraith Laboratories, Inc,,

Knoxville, Tennessee. Nitrogen analyses were performed by the

Kjeldahl method.








Apparatus



It 1vs necessary to exclude the presence of oxygen from

reactant solutions used in kinetic studies due to the extreme

sensitivity of the chromium(II) species to oxidation in solution.

This was accomplished by deaeration of all solutions using three-

neck flasks fitted with nitrogen inlet, outlet, and rubber scptum

for syrir.ge withdrawal of solutions. Times for purging ranged

from thirtL minutes to one hour. Transfers of deaerated solutions

--r ... de: v ?- r 1-'n'.?s rrreduated syringes fitted with s+.3inless

srtel nledicc,

For all kinetic studies at least one of the reactants exhi-

bited a characteristic absorbance in tne visible and/or near ultra-

violet region. Reactions were then monitored optically at the

respective absorbances.

Kinetic St dies

Fast reactions (1 msec < ti > 15 sec) were monitored using a

Durrun-Gib-son Stopped-Flow Spectrophotometer equipped with tungsten

and deuteriia; light sources, Kel-F flow system, and an ANINCO

4-8600 external ternerature bath. Use of an external circulating

pu op nain-i, nd a constant temperature environment for the drive

syringes, r.ixing jet, and observation chamber.

Thfe' ostatted, deaerated reactant solutions were transferred

t' +- ,f:r' n1 blhi' 20 nrl reservoir syringes. The

d:.:.: ..~' .ii front the reservoir marines ub a sterl

of o.-; ,, .. Sal ei-l valves. ilhc filled caive yl'in. s 'r' ere

.b:. .11.- A ;:' to tie c-sired t e' nature (esti'::ted)








for five to thirty minutes. A plunger armed by nitrogen pressure

(65 psi) and fitted with external, integrated trigger then forced

the roacta:nt solutions into the observation chamber via the mi.ring

jet. MKirxin and instrument mechanical dead tines (2 msec) were,

in some instnrices, comparable to the shortest half-life of the

reactions so studied. I'or these cases it was necessary to allow

for the ndadI txine and b -in actual reaction study after one or two

half-lives n d passed.

Data n ,-a record a first on the storage oscilloscope, tr.en

r": --7 ?, 3?000 I-ed Polaroid F-iL! (B & ) uLin-

a I'olaroid n C nkra bisn mounted on a Tektronix C-27 Oscillo-

scope Cancera with appropriate bozel attachment. Kinetic data :ere

obtsijncir ii; tL teprr.rature range 15.00C to 45.01C as necessary,

No special precautions were tazkn to exclude o:ygen other

than flushin-, the flow system with several volumes of the deacrated

reactant solutions. This technique was found to be sufficient for

solutions of cihromium(II) ion concentration greater than 4 x 104

I as evidenced by reproducibility of the observed rate constant

within n 10;;) frr the Cr(II)-[Co(en)2(OOCCH(CH 3)S)]+ reaction when

the concentr'ticn cf Cr(II) prior to mixing was 4.3 x 10 14.

r'elow this -.vel partial consumption of the reductant occurred.

Slow reactions were followed by use of Cary 14 Recording

Snrcfrorhotal{et.er fitted with constant temperature cell housing.

S:'-r;s 1 :In'.-i *':d with an /-iI:'CO 4-8605 constant terpcra-

S.'i ti, ..:; y ",-t'- c:ternal circulation; p p. Reaction

ueu. orn i.






bome of the reactions followed on the Cary 11 utilized a

glass rizring apparatus described pictorially elsewhere.14b

Essentially the apparatus consisted of two reservoirs placed so

as to form a "V" with an aperture for a rubber septum stopper to

introduce the respective reactant solution and with provision for

maintaining an inert atmosphere. At the apex of and perpendicular

to the "V" was delivery tube with pressure equalizing arm, fitted

with ground glass joint appropriate for the quartz cells. Tilting

the app-ratus, and shaking to mix, then draining into the cell

effectpd reaction. Initial data were obtainable within ten seconds

of ix

S ectroohotometry

Visible and near-ultrav-iolet spectra were obtained with a

Cary 'odel 1A4 Recording Spectrophotometer. Sample solutions and

baselines of pure solvent, were run against air as the reference.

Quartz cells of 0.10, 1.00, 2.00, 5.00, and 10.0 cm path lengths

were available.

Infrared spectra were obtained from 4,000 to 625 cnm- with

a Perkin-Elmer Iodel 337 recording infrared spectrophotometer.

All sar.ples were run as KBr pellets.



I'Pi spectra a re obtained with a Varian A-60A Analytical NM'

Spectrophotomcter with a magnet temperature of 370C. In order to

reach corccnttion of c o--ex larre enough to obtain r-:ani. -" 1

srFc- fC t r ? -rc dto-nro'oniic e-d c- plex, it was necessary

o ;- rch aie tdo e chloride salt.

i *::c actio o 0.300 Tr:ole of co.plcx








with 0.0'5 nnole of tetraphenylarsonium chloride (G. F. S.-ith) in

a minirun volume of water. The resulting tetraphenylarsonium

perchloratc formed was removed by filtration, and the filtrate

evaporated to dryness on a rotary evaporator and dried in a vacuum

desiccator overnight.

Spectra of saturated solutions of the chloride in deuterium

oxide (9'X.5^, Mathescn, Colerian, and Bell) were obtained with Ti-S

used as an external standard. Solutions were made acidic (pD = 1)

with trifluoroacetic anhydride (Aldrich).

Th solubility of the lactic acid and methylthioacetic acid

conDle-. -osed no such problem and spectra vere obtained using

the perc'ilorate salt.



All ion exchange separations were carried out using Biorad

AG 50'!-)12, 200-400 nesh (purified standard Dowex resin of the same

designation), analytical grade cation exchange resin. The resin

was converted to either the sodium or lithium form from the hydro-

gen fcr-. by soaking and washing the resin in a solution of 1 IN in

NaOH--I1.CIO1 or LiOH--LiC104, then rinsing with distilled water

until the eluent was neutral to Hydrion paper.

Dec:ceuse of the fine mesh of the resin, nitrogen pressure

(135 psi) ias used to increase the flow rate from 1-2 ml/min to

6-7? r./min. Separation of bands was adequate under conditions of

the rl-" flow rat". Elution characteristics for several ions of

intre-c: ae -iven in Table I (p. 271.







Treatment of Kinetic Data



For all the oxidation-reduction reactions studied it was

determined that the reactions were first order in both oxidant and

reducLant over the ranges considered, The stoichiometry was

established to be one nole of oxidant consumed per mole of reduc-

tart, vi-(e infra. The differential rate law then for reactions

first order in each species may be expressed as follows:

-<[]t0 [o]
kb = obs[R][]
d dL

where [Rl and [0] are the respective concentrations of reductant

and oxij Z nt and k is the rate constant observed at a given

acidity. In some cases the rate was also a function of acidity

as '-?l. rter rticn of acid dependencies will be discussed rith

the specific reactions involved.

Integration of the differential equation above yields, when

the reductant is in excess,

[R], x ([R]o [O]o)kt [R]o
og ( ) = -- + log ( ) (2)
[o]o x 2.303 [O]o

where the subscriot (o) represents initial concentrations and x

represents the concentration consumed at time, t.

The use of Beer's Law (A = ebC, where the terms are absor-

bance, ~~~-ir exti iclon coefficient, cell path r, i.' in cm, and

,or: 'icn in .. 1,1>- eer liter. rCspectively) allow1i0 Uie e:rqys-

;ion c. iaion (.-) in terij of experincntall obcsrvable para-

rcc :.o '. one r rI e of the reactants and/or products has a








characteristic absorbance at the wave length monitored, equation

(2) nay be expressed as


[R]o [R]o
A + ( -- l)Ao ( )A-
t + []o [o]o
log -
At A


( [Hao [ol)kt [R]o
S+ g ( -- ) (3)
2.303 [0]o

where A is the absorbance observed at the respective times (o, t,

and C ). plot of the left side of equation ()) vs tiie allows

an observert rate constant to be determined (i.e., slope =

([lo [o I)k

2.303

vr ~- -r froible a large excess of red-,ctant ([R, > 10[01]o

v-s used. Under this condition equation (3) may then be reduced

to the forn

k t
kobs
log (At A ) = +2.3 log (A, A. ) (4)

where ko = [R]ok. In this case a plot of log (At A) vs t

allows an observed rate constant to be obtained. Use of the

average va.juo of []o reduces the inherent error in the approxi-

mation leorii'rg to equation (4). The derivation of equations (3)
10
S(1') c" fcuj; ( 1G' o:nhere,0

o" < coses v- the for la rd and reverse rate constants

i to i ,.Jni. oi q equi.ilibriu reaction of the type


+ HI' -
1"r








here [FPo = 0, [Zlo = Z, and hydroTgn ion was present in large

excess, equation (4) may be utilized by replacing A. with A.
eTus
Thus


log (A -A ) =
t eq

where kobs = kf + kr The

were thus derived from


k
kobs _
f 1 + (1/K H]) r


Elucidation of tis metnod


k t
-obs + leg (Ao )
2.303 eq

forward and reverse rate constants




kobs [P eq

1 + K eq e [Z 00E 17

is discussed elsewhere.3031


Evaluation of Activation Parameters


Consider the simple binolecular reaction

K* k*
R + 0 (RO)* t Products

where species R and 0 form the activated complex (RO)* which then

may proceed to products. Following methods previously outlined,3

K* can be treated as an equilibrium constant and the reaction rate

expressed as


-dh]
at


kT
S I ".T''1 = k'*K"*(R.I-C; l = 1'4'RliO0]


1k =


CA I I .n1


nl's c I :tsani,








h is Planck's constant, and T is the absolute temperature. Treat-

nent of K* as an equilibri.n con:ttant alloue the expression


(;G* = -RT In K*


to be used. Then


= hiT K-/i kT AS/R e-'H*/RT
k h e h t f / e


Eqnnati-on (0) may then

euiAi.on to yield


be co:-c r Led into the form of a linear


k

lg ( log ( + 2. 2.RT
T h 2,.3R 23T


k
A plot of log (.-) vs

slops, ci:


then yields a value for AH* from the
T


AH* = -2.3R(m)


The entropy of activation can be obtained from the y-intercept, b,

by using equation (12).


AS* = b log ( )2.3R
11














RESULTS


Characterization of Comolexes



Spectral and ion exchange properties upon which the charac-

terization of all complexes are partially based are summarized in

Tablrc. II anJ III. Pr:'on rs.-iec r onanco data for the

methyl, m.thylene and nethinyi protons of interest are listed in

Table IV.

2-. rz.=,.tcrrop :bi-r t' ,1..r.c h:- ar ',': t. ] ,''_i. -m
[Co(en) (OOCCH(CIH,)S)JC0l

Characterization of this complex was accomplished by elemen-

tal analysis, infrared, visible, ultraviolet, and proton magnetic

resonance spectroscopy, ion exchange chromotography, acidity

studies and reaction patterns both independently and in comparison

with the previously reported mercaptoacetabo complex.0 The basis

for the formulation of the complex as -.ritten, bound through car-

boxylate oxygen and mercaptide sulfur, will nowo be summarized.

Elemental analyses for carbon, hydrogen, nitrogen, sulfur

and cobalt as previously reported in the Exrerimental section were

in r e."' with i c-ilai'i rcent-cs. The infra-red
-1
spc, of no :::orptio e in tce 2,50 e.-

areu wiLcn suuMa.1.iaf.laL- i Sauur uL in deproLontucid in ith-L coa-

plx a sold, inte;- rpi- 160 i -1,-
p1~ : a V so 1d Lne ;q rpi] 1,60 adlI,)~ -le







TABLE I

Elution Characteristics of Some Cobalt(III) and Chromium(III) Speciesa


Species Eluentb Vl( i)c' V2(ml)d',e


cis-[Co(en) 2 (C1)2]) 1 65 40
[Co(en) (00CCH(CH )S)] 1 60 30
2 35 20
[Co(oen)2(oCcc;(CH O))o)+ 1 55 30
FCo(en)20OOCCH(CH 10) 2+ 2 100 50
LCo(c)o2 C'2 3"i 325 1;0
[Co(cn)2( (iCCHi2)] I 2 1 420 140
2 125 70
[Co(H20) '2+ 1 210 100
2 90 40
[Co(on)3 1 1000 --
2 600 --
[Cr(H20)4(COCCI!(CH )S)] 1 45 25
[Cr(H20) (OOCCH(CHl)0)] 1 50 25
[Cr(H20) (OOCCH(CH )SH)]2+ 2 90 40
[Cr(H2O) (OOCCH(CH )OH)] 2 2 85 40
[Cr(H~O) (OOCCH SCH )]2 1 115 50
[Cr(H2o) (OOCCH2NH3) 3] 2 550 200
[Cr(H20)6 + 1 230 100
2 110 45

a30 x 1.25 cm column in Na+ form. 0.2 rmmole used of each ion. bEluents:
(1) 1.0 N NaCIO,,; (2) 0.10 M HC10--2.0 Y NaC10,. cVolume eluent passed
before barKd ,iwrtl s to elute. CVoltune eluenL pass for elutio:n oi band.
e i :',.-r.t \' ;.plicate rnu:s varied up to 15%.
















Ci (d d ol dcq


S0 0 0 0 0 ON
cd r-H ri H r-H H -H r-H H \ 0 0 N


H I HI C'- H rH c- -- Ic
0 00 0 C)


0

I i
I I '-

Co
0o
C-]


'o
0
"3


Co
NL


0
I Hf
0
.2


+
+ + +
r. r'3 -

CO 0 C)


u y u
u u
U 000


I i
I i I


co
CO
IN
00
C


+ +
+ F ++r-
+ + c -



\2 \2 F--- I: U-t -'1d .'
+ +v + 1 N N h 4 r- 0



l N- C. ) CC NqL
CM CI 0 C- U- 0 N 0 C 0) *C. 0
C/ !-- D M C NC C. ) 0





S- 9N C) C, N



V . Cz) C.) ( N C ..... C ..
C CC C O0


1-1
C,
0 C


C.
U

o u|






-C



ni a
















dj V% io r-i CH o d coo 0 D 0 d 1
4 H ri H-l Hr l N


N N CO

'4-'
.-I r-1


N N t\ N co N
S I I I( I r- I C- C-
Hr1 H- H N


0
0
I I I I I


C
o
0



Ci
N
Ml


C-iI-- -
+ + +

i + i r N + + + + .
\ .- C rC, -1 + +
SN 0C. N
(7 CN \ Cq cn V ~
u0 r 0
L --u i: C -- N [__ :1 C 4 + IN I



C) C) (. C C-: Cp a (9 C)i: a.: C) 32 ) -*.





G 6 ( i o i) cii ci o oi o 3 o 6 ci







TABLE IV
PIPR of Various Protons Adjacent to Chalcogens


Species Medium Methinyla Methyla Methylenea Ref.
Proton Protons Protons


!o. .-"if OCH 3 )SSH
DOOCCH(CH, )SD
NaOOCCH(CFH )SD
[Co(en)2(C 0c (CH,)S )
CCol(n)2 (" C -3)S')
NaOOCCH(CH )OD
+
[Con)2 ::(! -+


CKa (SC)2C 3 C2

[Co(en)2 ( '.:. )

[Co(en)2(Cr C:; C.,T~r) )2+
1)2 1CH2COODb
1 )21Gkl 2 Cool) 2+
[Co(en)2 (CoOCUi 2iH) 2+
NaOOCCH SD
HOOCCH2SH
[Co(en)2 (OCCi2S) -
[Co(en)2(OOCCH2S)]+
NaOOCCH2OD
DOOCCH2OD
[Cc(en)2(OOCCH 2 O)
[Co(en) 2 (OOCC'20D)]2+


CoV n.(i- m t 'i y )*- "-

i)""nfi pcl ****' :***".;C .- (1


CDC1
D 20
2
D20

D20-CF3COOD
D20
D20
D20-CF3COOD
D20

CF COOD
D20
D 0-CF COOD
D20 3

D20



D20-CF COOD
DO
D20
D20
D 0 0
D 0-CF COOD
D 0
)12+ O


*-1.- o01. II. c'his .:ork.


3.58
3.65
3.45
3.51
3.51
4.09
3.2
4.36




















4.51
3.60


1.56
1.48
1.35
1.32
1.32
1.30
1.25
1.32
2.08


2.38
2.38














2.22
2.22


2.55
A=3.62

3.75
3.75
3.58
3.58
3.51
3.32.
3.10
3.10
3.99
4.29
3.90
4.26
3.27
3.09


~----~-~--------~----


).
I).


Willinms ; il U. H. iuch, J. A.:r. "-. Soc., S L. ,








indicate the presence of coordinated carboxylate. 1633 These

data are consistent with the ion exchange studies which indicated

a unipositivo ion (Table I).

Spectral parameters in the visible and near-ultraviolet for

the compound are as follows: [h(E): 517(152 t 2), 360(340 20),

282(1.23 t 0.02 x 10 )]. The error associated with the molar

absorptivity at 360 nm corresponding to the 2 IAl d-d trans-

ition is due to its super position on the tail of the intense tbnd

at 2S2 nr. This contributes to the rather high value for the molar

absnortivitv at 60 nn 3:ith intensity borrowi-ine front the higher

cni --j. peak ai u app; i important. it was necessary to use the

ultraviolet rather than the visible source because the slit vidth

vith the former allowed for better resolution at the 360 nm setting

to define the shoulder clearly.

The near-ultraviolet peak is assigned to sulfur-to-metal

charge transfer due to its large molar absorptivity and the appear-

ance of similar peaks (2 nm) in the spectra of the 2-mercaptoace-

tato, 2-mercaptoethanolato and 2-mercaptoethylamine complexes.

Similar peaks are absent in the spectra of the oxygen analogues.

A fourth absorbance with maximum near 220 nm and molar absorptivity

of 19.000 + 2,000 was noticed for the thiolactato, lactato, mercap-

toacetato, glycolato and glycinato species, but reproducibility is

less than desirable. It is tempting to assign this transition to

carboy't!ate o:'-:e-n-trc-rt 1 charrc transfer which would then corre-

lrt the low r:r r:r r" lf;:r-to-nmctal chr.ge transfer ridth the

gr.ir csc c oict: ci -Liur re l 'tivo t o:o0gcn.

Is for *It er;c.~ to co







near-ultraviolet spectral parameters over the pH range 0-7 lead

to the conclusion that the thiclactato complex is not detectably

basic under the conditions studied. Kinetic data support this

observation, vide infra. The lack of basicity in this range is

comparable to the results observed for the mercaptoacetato cobalt

complex10 and thiolochromiui complex. '35

Proton magnetic resonance spectra of several cobalt(III)-

chalcoganide complexes previously have shown that the resonance of

the methylene protons contiguous to the coordinated chalcogenide is

shifted upfield by 0.2 o.7 ppm from the neutral free ligands and

that deuteronation of coordinated alko:nide shifts the methylene

resonance to near that of the free ligand.1019'36 It has also been

demonstrated that the resonance of ethylene protons adjacent to

coordinated sulfur remains shifted equally upfield from the free

ligand in both neutral and acidic media.10,19

From these results it was expected that the methinyl proton

on the propionate skeleton would exhibit similar behavior. The

resonance of the methinyl proton adjacent to coordinated sulfur in

the 2-mercaptopropionate skeleton is shifted upfield from the neu-

tral free ligand resonance but by the smallest amount yet observed,

0.14 ppm. Such a small shift seems inadequate to diagnose sulfur

coordination except in combination with other supportive data such

as that presented here. The methyl resonance is similarly shifted

upfield by only a small amount, 0.16 ppm. The lack of response in

the methyl resonance to acidification supports the contention that

sulfur is not protonated over the acid range studied. The reacti-

vity patterns and product studies described below provide further

confirmation of the formulation of the complex as described.








Ltactatobis (athylenedianrine) cobalt(IlI) Perchlorate--[Co( n)2(0OCCH-

(CHi3)0) ]C10

Similar techniques were employed for the characterization of

this complex. Elemental analyses for carbon, hydrogen, nitrogen

and cobalt were in good agreement with the calculated values for

the complex as a unipositive ion coordinated through both carboxy-

late and alkoxide oxygen atoms (Experimental section). Further,

the complex eluted as a +1 ion with neutral fluent and as a +2 ion

with an eluent of pH = 1 using the previously described ion exchange

macLnous (ulO I),

Infrared spectral] data confirm the presence of coordinated

carboxylate which exhibits maxima at 1,630 cmr- and 1,350 cm-.

Determination of the: presence or absence of coordinated alkoxide

by examin-:tion of the 0-H stretching region was rendered impossible

due to broad N-H and C-H stretching mode absorption originating from

the ethylenediamine ligands in the 3,500 cm- and 3,000 cm-1

region.

Spectral parameters in the visible and near-ultraviolet re-

gions are as follows: [\(M): 517(138, 360(153) in neutral water

and 499(113), 349(123) in 0.10 1M HC10 ]. An absorption near 220 ran

with unreliable reproducibility in molar absorptivity was also

present.

c'e-ri.rat.on onf .The K of the cc~olcr wnI's necessary due to
3 a
its incl;,io n in tho rate cxpression at low p0l, viae infra. A value

ias obtained usin,: tlhr r.ethods of detenidnation which gave good

agreement. A pKa of 3.36 at 1.0 M ionic strenFth (LiCl10) was

obtainrl r- s pctrcal i s. Direct clc ctro-tric titration







with 0.100 N HC1 of one millimole of complex in 100 ml of water

gave pKa = 3.33. A titration with 0.100 N NaOH of one millimole

dissolved in 25.0 ml 0.100 N HC1 (providing a known excess)

yielded Ca = 3.43 ( /X = 0.10 M). A value of 3.37 t 0.06 was

adopted as the pKa of the complex. It can be seen that there is

little, if any, variation with ionic strength as has been found
37
elsewhere. The value of 3.37 is in reasonable comparison with

those for the analogous glycollato complex (pKa = 3.3 0.3) and

the 2-aninoethanol complex (pa = 3.59).37 The [Co(en),(OOCCH(Ch3)-

OH) |2; i is mor- rcid5c by a factor of 102.' and 102.2 over the

c(... ci; and [Co(iH, ) (HOCH ) J+ complexes with

pKa values of 6.1 and 5.58, respectively.3835

The pmr spectrum of this complex reveals a resonance for the

methyl group adjacent to alkoxide oxygen which is only slightly

shifted relative to the free ligand monoanion in both neutral and

acidic media (Table IV). In contrast, the methinyl proton reso-

nance is shifted upfield by 0.17 ppm from the free ligand mono-

anion. The comparisons evident in Table IV suggest that the shift

from the free neutral ligand will be comparable to that observed

for the glycollate complex. Together with the observation that

deuteronation effects a 0.44 por shift downfield compared to a

0.36 ppn shift for the glycollate complex, this appears to vali-

date the resonance shift criterion for alkoxide coordination.

e dii -' '"thyd sersitivity of the 2-mercaptocropionate metLinyl

I-,r;i' coordination is not -'c.r..ood but mar








The pmr spectra, the acidity of the complex, failure to

incorporate water on recrystallization and the striking similarity

in the visible spectra (both in acidic and neutral media) to the

analogous glycollato complex previously characterized and reported

provide evidence that the alcohol function is coordinated to the

metal center in solution. Reactivity patterns and product studies

described below are consistent with this conclusion. Further,

prolonged exposure of the complex to 0.10 M HC104 effects a first-
"7 -1
order reaction (k = 1.9 x 10 sec ) to what is assumed to be

c :)~' ( \' rC(~' C ) 2C:)]2+ on theU basis of srectral com:par-

ison t tthe cis-acouato-aquo and cis-,-c : .i -to-aquo analogues

(Table II).


[Co (rn)., (OOCC,-2SC. 3) J(C104)2

This cou.pojnd was characterized utilizing the methods employ-

ed for the previous compounds. Elemental analyses for carbon,

hydrogen, nitrogen and sulfur were found to be in good agreement

with the calculated values (Experimental section). The finding

that analyses for hydrogen and sulfur were very slightly lower and

higher, respectively, than the calculated values weighs heavily in

favor of coordination through carboxylate crygen and thioether

sulfur since the principal probable impurity, that with water

cnn(iirnit-d in pace of the thioether function, would reverse the

c:'' on cchanri cirocteristi.cs for the cormlcx were that

oi t *- ion ( < io I).

iii'rarIC byucLral uata Cluii'in tii presence of coordinated

crc 'n at -o c- and 1,350 c-
cnr .int, ve n x l. r a ': r e- onS at l,t'O cm- and 1.,350 c'-







16,33
which are characteristic of this mode of coordination. The

definition of sharp absorbances at 3,470 cm-1 and 3,400 cm1 char-

acteristic of N-H stretching modes argues against presence of

water in the solid since presence of the latter usually obscures

the N-H stretches.33

Spectral parameters in the visible and near-ultraviolet for

the compound are as follows: [N(M): 499(168), 360(250), 280

(7,300)]. As previously noted for the 2-mercaptopropionatocobalt-

(III) complex, the absorption at 280 nm is taken empirically to be

characteristic of metal-sulfur coordination. This transition

energy is attributable to alternative formulation only with exces-

sive difficulty. In the pH range 1-6 there was no change in the

spectral observations as would be expected for a weakly basic tri-

valeit sulfur atom. For the experimental conditions used in this

study the complex was found to be stable in solution for at least

four hours, a period much longer than that employed in the kinetic

measurements.

The advantage of using as a structural probe, the reaction of

a species with a reagent whose reactivity patterns are thoroughly

characterized, is demonstrated convincingly in this case. Reactions

of the complex with [Cr(H20)]2+ proceeds in two or three observable

steps to yield an isolable carboxylatopentaaquochromium(III) product

with the rate of the first step being independent of acid concentra-

tion, vide infra. This behavior cannot be reconciled with that

expected for the only apparent alternative formulation, [(en)2Co(H20)-

(0OCCHPSCH,)], which should be reduced in one step to the carbox-

ylatopentaaquochromium(III) product, or with an inverse acid







concentration dependence to yiold [Cr(H20)6 ]9, [Cr(H20)5(OOCC-

H2SCI3)2+ or [Cr(H2o)5(CHI: SC:2COOil". The last complex is, in

fact, believed to be the initial product which can convert to the

carboxylatopcntaaquochronium(III) product in two stops, chelation

by carboxylate followed by decholation at sulfur. However, its

genesis by a rapid, acid independent reaction cannot reasonably

be attributed to reaction with the alternative complex. Its

generation can be rationalized in terms of the formulation pre-

sented, vid- irfra.

The proton mametic resonance spectrum of the complex anoears

to provide corroborative support for the thioethcr function being

coordinated. The methyl resonance undergoes a shift of 0.30 ppm

downfield, relative to the esterified methylthiopropionic acid,

while the methyl resonance for the complex of S-methylcysteine in

wnicn tne thloetner is not coordinated' remains unshiftea relative

to the free ligand value. Further, the methylene resonance is

shifted downfield by 0.13 ppm for this complex, relative to a

comparable ligand, whereas for the pendant thioether complex, the

shift is 0,2 ppm upfield. Thus, if the complex is regarded as

being derived from the mercaptoacetate complex by a methyl car-

bonium ion substitution on coordinated sulfur, the effect on the

methyl and methylene resonances are similar to but larger than

that arising from deuteronation of coordinated alkoxide, vide infra.

Anrittedly, this rerrocents an entirely empirical approach to the

interpretation of pmr shifts on coordination which must be rcoar-

( as .ri-. 'C anid ; jct to r c'%aluation a; evidence accumu-

lntes. For unately, hr ssi nr :nt of thiocther as an active







donor function in this complex rests as well on the broader

evidence adduced previously.

Glycinatobis(ethylenediarine)cobalt(III) Perchlorate--[Co(en)2(00-

CCH2 1:H2)](C104)2

This compound was characterized by elemental analysis, visible

and infrared speciroccopy, and ion exchange chromatography. Elemen-

tal analyses for carbon, hydrogen, nitrogen and cobalt were in good

agreement with the calculated values (Experimental section). Ion

exchange behavior identified the ion as a +2 species with both

n~outra: cd 0.10: rcid ;luCnt (Table I).

Visible spectral parameters for the complex in rater and

0.10 1 acid are as follows: [,\(): 487(98), 346(106)]. The

infrared spectrum exhibited intense peaks at 1,640 cm-1 and 1,340
-1 16,33
cm characteristic of coordinated carboxylate. 3 The spectral

parameters are in good agreement with the previously prepared and

reported chloride salt of the complex, [Co(en)2(OOCCH2NKH2)C12.16

2-Mvercaptonropiona.toi)ntaaouochromium (III) Ion (Mercaptide-Bound)--
[Cr(': O)(SCv(C:H )CcOS)]+
!-3

The reaction of aqueous chromium(II) with [Co(en)2(OOCCH-

(CH )S)+ is extremely rapid and results in >90A incorporation of

the rercatopropionate in the chrodrumn(III) product coordination

sphere, vid-" infra, On the basis of these observations and the

hi rh ster(-e irrohabiilii! of a doubly. bridr"' rechanisn utilizing

1li c- u-s': p ]::rctr" Sulfuz evident in models

le.i tc .': :l i i the i'sC product of the oxidaoion-

rtuction ~, ) .. ('. )Cju ) ( When the reaction of the

cnltT.I i) Z.Tri;" hrori-;.(I) ion is -;od il a slight







excess of oxidant a. fleectiig intermediate is observed which wuder-
-2 -1 -1
goes a rapid subsequent reaction (kobs = 3.29 x 10 M sec- at

[H] = 0.90 1, see kinetic description below) to produce a species

spectrally identical to the chelated 2-nercaptopropionatotetra-

aquochromiuv(III) ion, a sufficiently long-lived species to be

partially characterized in solution and the only isolable product

of the reaction of the cobalt(III) complex in the presence of the

excess chromini(II) ion.

The sulfur-bound ronodentate intomrediate is not sufficiently

nor"i-lived t+ c'dter:ne its spectral parameters, but the relatively

smAll chaL~ne i n aibsor nce observed on conversion to the chelated

product suggests ananomalously high molar absorptivity for the

intermediate. Since the intermediate, as formulated, is comparable

to the [Cr(H20)5SH]2- species (see Table III), the spectral obser-

vations are then reasncrable and lend credence to the nature of the

short-lived species as proposed.

2-lercaptopropionatotetraaqucchronium (III) Ion--[Cr(H20)4(00CCH-

(CH )S)]+

The isolable product obtained from the previously discussed

species or from reaction of the cnbalt(III) complex with excess

chrc:miun(II) is so for;.;ulatcd on tle basis of ion-exchange behav-

ior and spectral parameters observed from the products of the

rrretion of FrC(- (pnCCHiCH S)) and Cr(II), a representative



A 210 1 reaction: r -.xturl initit 1 5.0 x 10-3 M in Co(l),

5.1 x i0-5 M in Cr(ll) and 0.020 1. in h was exposed to the air

aI a 'i.-.on p1rio nc. chlari-rd onto a 30 On




xml version 1.0 standalone yes
Volume_Errors
Errors
PageID P74
ErrorID 4
P76
4
P134
4
P146
4
P180
4
P182
4






(1.2 cm diiieter) column of Biorad-purified Dowex 50 w-x2, 200,

400 )resh, cation exchange resin in the lithium form. After a

charring and washing time of about thirty minutes a +1 chromium

species was eluted with 0.25 I! LiCIO The band began to elute

after passing about 50 ml of fluent and was collected in about

35 :.l, Volume was adjusted to 100 ml with eluent serving as

diluent. Aliquots of this solution were then used for further

study with appropriate additions of standard !C1004 and solid

LiClO4 to maintain the desired pH and ionic strength. Separate

r-- r..... c'' ) O-95a- rTr'e) y of the 2-rercaptorenionate-sonain-

int chrecrii:'.(ili) ion, oi a 1:1 mole basis relative to the amount

of cobalLt(III) complex used initially, based on the previously

described ncthod for chromium determination, vide supra.

ThaL this product is the chelated 2--mercaptopropionate com-

plex is confirncd by the +1 charge and its spectral parameters

[A(E): 545(71.2 1,0), 440(52.2 1.0), 265(5040 150)] (the

uncertainties in the molar absorptivities are due to subsequent

reaction of the species, vide infra). These values compare

favorably ilth those of rmercaptoacetatotetraaquochromium(III) ion

and thiolopentaaauochromiuzm(III) ion (Table III). In both of the

latter species tne hifh energy d-d transition (440 nm) and the

near-ultraviolet absorbance (265 nm, provisionally assigned to

sulfur-to-metal charge transfer. video cuo1ra, strongly suggest

r ;,tido cairijic. The fact Ithai the low e crV 6y d-d transi-

i i -) ocrrs i'or b1ot acid co: lcxes ard is blue-shiftid

iui, ie i. luc. or hU Lilo ani lqaj c.iUroiLum species indicates

c;:::-. c-rinati (:1 e 111). Sinlar a ob:rvations have








been reported for several carboxylate-bound chromnium(III)

species.35 The high molar absorptivities observed suggest

chelation, which would lower the overall symmetry. Examination

of the miolir absorptivities of several cis-aquo carboxylate com-

plexes of both cobalt(III) and chiromium(III) further substan-

tiate this explanation (Table II and III). Occurence of the

sulfur-to-metal charge transfer absorbance some 20 nm higher in

energy for chromium(III) complexes than for cobalt(IIl) complexes

lends credence to the assignation of the transition to the charge

tra-,:fcr I n. vie-: of : :ocr e tendency of chro-.in'(lll) to Ie

reducca relative to that of caobalt(ll), (Ed (13+/2+): +1.84 v

(Co3/Co2+), -0.41 v (Cr+/Cr2+)).40

2-'oercaptopropionatopentaaquochromium(III) Ion (Carboxr.ate-
Bound)--LCr(H20) (OOCCH(CH )SH)] _

Isolation in solution of this species from the direct

reaction of the cobalt(II) complex and chromium(II) is not possible

due to presence of the cobalt(II) ion produced. Elution character-

istics of the product +2 ions are nearly the same,resulting in an

ineffectual separation. However, isolation of the previously

described chromium(III) chelate complex followed by subsequent reac-

tion in various acid solutions affords a route to an equilibrium

mixture of the chelate and pendant (carboxylate-bound) chromium(III)

ions, Thin mixture c,(n then be separated by ion exchange. Use of

the ionic rlnt- n -th ran; 0,25 ( I to 0.50 1' in HCIO4--LiCIO4

f'o" the C aii.libriun r,,Nwction r-nlts in thef cholate fraction not

being retajnei on the column unile the +2 monodentate form remains.

P:lution v' 1.0 1 I.i.l alr-vs isolation of tle ronc nintlate fonn.







Eecr.use of subsequent reaction of the monodentate carboxylate

ion to hexa:quochromium(III) it is not possible to recover total-

ly the initial amount of chromium as either chelate or monodentate.

The monodentate product was characterized by its +2 chargee

and the similarity of its visible spectrum FA(E): 568(25.0 t 1.0).

411(24.5 t 1.0)] to previously prepared carboxylatochromiumr(III)

ion.35 Again a high energy peak at about 210 nm was present, but

again o- varying intensity (E = 19,000 t 2,000).

Upon isolation, solutions of the nonodentate carboxylato

pr tci can be observed to undergo subsequent reaction spectro-

p l -iTcall'. Ion exc! car.e separation of an equilibrium mix-

ture after several days results in three separable fractions. The

firrt fraction :lutes as a +1 ion and is spectrally identical tc

the original LGr(H20) (OCCCi(CH )S)] ion whereas the second frac-

tion is the monouentate species, the +2 ion LCr(nO2)5(i A;Ci(CGh3)-

SH) 2+. The presence of the chelate form shows that dechelation

of the chroniun-sulfur bond is reversible. The third fraction was

identified as [Cr(H20)63+.

La.cLaftopentaaquochromiuim(II) Ion (Alkoxide-Bound)--[Cr(H20) (OC: -

(CH, )cooH) i

That this species is the first product formed in the reaction

between aqueous chromium(II) ion and [Co(en)2(OOCCH(CKH)OH)]2+ at

pH = 3.4 (there the alcohol function is substantially deprotorated)

or- c" te p'- .!i i.,nely t prooctionalto ? roen ion concentration

ati .. 'o lo.iaJ c I orison c.ith le

suI.IU iail0(,;,. iiie ion lias not ;Lu(n inucpandciiLly ideriti'iyd,

\w-. "







Lactatop3ntanquochroTLium(ITIl) Ion (Carboxylate-Bound)--[Cr(H120)

(oocc:;(cf ) i()_O

From the reaction of [Co(en) (OOCCH(CHi)OH)]+ (1.0 x 102
-2,
i), with chror.ium(II) (1.1 x 10-2 M) in 0.100 A HC10,, a diposi-
tive chromium(iIl) product, characterized as the monodentate 3acta-

to complex, can be isolated using ion exchange techniques. The

visible spectrum in 0.10 U HC104 had the following parameters:

[X()): 568(26.6 1.0), 412(33.2 1 1.0)]. The molar absorptivi-

ties are slightly higher than previously reported values of 25.7

ra. y.2o fc' ...r.s r D'.. s, .-s41 ei it should b noted tcat

tic coirplex was not isolated in the previous work and values are

based on 35' complexation in a solution of lactate and chromium(III)

ion. It should be noted further that the peak ratios reported pre-

viously (CA /6E, = 0.82) do agree reasonably well irith values cb-

served in this work, (EA /E^2 = 0.81).

VTWen the reaction of [Co(en)2(00CCH(CH )OH)]2+ was perfonred

at an initial acidity of 4.0 x 10 14 HCIO two chromium(III)-con-

taining complexes of the lactate were eluted from the cation ex-

change column in the approximate molar ratio of 2:1. The first and

major fraction was eluted with neutral eluent as a pink +1 ion

describ-d below. The second fraction eluted with the sarae neutral

eluent as a +2 ion that changed over short but observable periods

fron a blue color on the column to a pink color in solution with a

r i;: iC' -):!oal to i-ht o~' the first fraction (pi = 2.9). The

l pcci. Le thle Iko, ide-bound nonodcntated complex but its

il-, g exio.cCe mc, ii. chara tcerization le ntLative. Acidifi-

ca/or of bomtl fmctio:io i' 1 l~,e the spectri,- of the monodcn-







tate species within the time span necessary for the experiment

(5-10 min).

From the fact that the +2 and +1 ions are readily inter-

convertible at a measurable rate, vide infra, it can be surmised

that the +1 ion is not the [Cr(H20)4(OH)(OOCCH(CH 13)OH)]1 complex

for two reasons: (1) The pKa of the carboxylate-bound +2 lactate

complex would be expected to be near that of the acetate analogue,

[Cr(H2 )5(OOCCH3)]21, that is, pKa = 4..525 (2) If the hydroxy

complex were important, its fomration would be expected to be

dif~C^ c" controlled.

Lact:C;, rauociio (II) lon--FCr(- (0Cc (O (C:i )0)+

This chelated species vwas concluded to be the major product

observed (~65',) from the reaction of [Co(en)2(OOCCH(CH3)O)]+ and

chro-iun(Il) at pH = 3,4. Cation exchange separation characterized

the sncies as a +1 ion which had spectral parameters of [A(E):

548(31.2), 437(38.1)] at pH = 2,8. Subjecting the monodentate

carboxylato species to pH = 2.8 resulted in a change producing sin-

ilar spectral parameters based on total chromium(III), [N():

548(31 2), 437(37 2)], within the five minutes necessary for

manip.l lotion.

U anination of the spectral paran ters of the proposed che-

late cor-plex and comparison to the analogous mercaptopropionate

cor^?"1 i'i c5 n,"*r r- rcire s'lidi, fcr-tulted, sur'aorts the chielnte

for Lh +1 c ..,i: i nrgics for the d-d I -sitions a-e

cs i i c:cl for hi the alkoxide and merca tide complexes

of chro- i m(1l1) ( able III) as is found to be the case for the

cob-'.c(''.) c 'l - .....- (5leo- ]* ). 'nie rr-la:iftinru o the which







energy band and blue-shifting of the low energy band relative

to chromiur(Il) occur in the regions expected for coordination

through alkoxide and carboxylate but not expected for coordina-

tion of either function alone, Finally, the observed molar ab-

sorptivity would be anomalously high for chromium(III) complexes

bound throlrh carboxylate alone in comparison to similar com-

plexes.35
1?' "*i''*. '" i. r i.A l,:,I i *1 : 'i I-Ii I -Ii'TTT 'i -i.r .. ITr ii i."I -, c:_r-1;...-. i-



By cor-r'rison to thc rmrcaptide corolex, this species is

cxo-ctcd to L- the first pr:ouct for.im;c in the reaction between

aqueous chromiun(lI) ion and [Co(en)2(OOCCH2SCH3)]2+. Since the

ion has not been independently identified and its formulation is

based on kinetic results, discussion of it will be deferred to

the appropriate section.

Methylthioacotatotetraaquochroinium(III) Ion--[Cr(H12O0)4(00CCH2SC-

H^

Formulation of this ion as the product of a subsequent

reaction of the thioether-bound monodentate species described above

is based on kinetic results and will therefore be discussed in a

laotr section.

Methylthioacetatopentaa iochrorium III) Ion (Carboxylate-Bound)--
.-. 2

Thi:. e;:ccis C:i icolated in solution in 'o(Y yield usinr

ae: : on cxc!:i e t(c:nrniqcios from the reaction of [Co(en)2((C

HBy I wiith chromium(II) ion with the latter in deficiency, equi-

v ,-i cy or cs. The rcies is for-,.latLd r +2 ion on the







basin of ion exchange elution behavior (Table II). The visible

spectrum in 0.50 I LIiCiO4, pH = 3.4,exhibited the following para-

nreters: [A(): 567(26.7), 412(25.9)]. The visible spectrum was

found to be quite similar to that of the [Cr(H20)5(OOCCH2S)]21+ and

other carboxylato-chromium(III) species (Table III), thus confirming

formulation of the ion as described. The +2 ion is the only

obser-vable product of the reaction of [Co(en)2(OOCCH2SCHi )]+

and Cr(li) at acid concentrations in the range 0.10 M to 0,10 M.

Glyci;; Lopen Laaquochromium(III) Ion (Carboxylate-Bound)--[Cr(H2 C) -



This species can be isolated in solution in > 90,, yield using

cation exchange techniques from the reaction of [Co(en)2(OOCCH2-

I'H2 ))J with chroiTiium(II) with the latter in deficiency, equiva-
lern- or excess, The blue product is formulated as a +3 ion on

the basis of ion exchange elution behavior (Table I). The visible

spectrum in 1.0 M1 LiC104, pH = 3.5 exhibited the following para-

meters: [A(E): 573(22.0 + 1.0), 412(23.0 1.6)]. Although the

peak positions are quite similar to [Cr(H20)6]3 the molar

absorptivities eliminate the possibility of this complex as a
pro+dct (Table III). Spectral comparison to other carboxylato-

chri i>,(III) p cice confirms formulation of the ion as described.

The +3 ion is the only observable product of the reaction of
[C;(, "-I (COC:',"" -)"12+ and Cr-!I) at acid co:centrations in the

fra,' t. 4.1 taO 1.0 x 10-' .
oten-LCr( )( )

o aI s-, :; 'of U., ':: Lo 4.5 of a solution of the







above, the solution color chariyed rapidly from blue to green, but

the naxima of the visible absorption remained essentially constant.

Letting the solution equilibra.le for several days resulted in the

solution acquiring a red-violet hue. The higher energy absorption

remained very nearly at its previous position, but the low energy

absorption shifted to 560 5 ni. On passing the equilibrated

solution through a cation exchange column in the lithium form, it

was noticed that a faintly colored fraction was not retained on

the column, probably due to the high ionic strength (1.0 u) of

th" cher','r solution. Coll].' ~l of tlis fraction -ncd spectral

analysis yielded the -ol.lowii,-: .E>(e): 55i(3S.U 0 5), 420(41.0

t 6)]. From the high molar absorptivities compared to carboxy-

late-bound chromiurm(ill) species but relatively normal for

cholate species (Table III), the spectral shift of the low energy

band to higher energy and the foct that the ion was not retained

on the cation exchange column, it is tentatively concluded that

the species is the chelated [Cr(H20)4(OOCCH22l)]2+ ion.



Rcatinn of C^~cJCm(1T) i.ti th. Cfb-._lt(III) Complexes



2-M-ercagitonropionato);l s (ethyln ondJ.iai)cphgcltIII_ _on--
[Co(on) (ooccH(cH2)S ))]

The stoichiometry of this reaction was determined by ion

cxchanre r' c!ration of the rc'e icn prcAi:3s follow d bt- chromiu-

an;,-:i of the appuro, riac f -,-"iri : r reaction w~ere

cA vsi. o o.in) .: i .:tti. i J in excisO.

A sohvt-. of chrc i:: .) i l J i c luiin of the








complex with the usual exclusion of air. Within fifteen minutes

the r action mixture was exposed to the air and charged onto a

sui!-tbly prepared column. In each case the amount of chromium(III)

product elutable as a +1 ion closely approximated that of the

deficient reactant (Table V). This is indicative that the

oxidant and reluctant react in equimolar amounts. This first

isolable product was characterized as the chelated chronium(III)

product, [Cr(2,0)4(OOCCI(CH3)S)]+, vide supra. The fact that

the ligand is -100i incorporated in the product at varying reac-

tant ratios is -t lon to er halish tc- reaction as i:r-srer'

in nature.2 As previously discussed, the chelated +i ion was

concluded to be a secondary product of the initial oxidation-

reduction reaction, the result of the ring closure of the primary

monodentate product, [Cr(HiO)5(S(CH )CHCOOH)]2+. The only other

chromium-containing fraction was obtained from reactions with

reluctant in excess and was characterized spectrally as the

air-oxidized dimer of chronium(III). 43 Spectra of the product

solutions corresponded within 4% to those calculated based on a

1:1 stoichiometry and sui;.aing the contributions of each species

remaining.

The rate of the Cr(TJ)-[Co(en)2(OCCH(CH )S)] reaction

was found to be measurable on the stopped-flow instrument. Since

the second-order rate constant was >105 (-"1 sec-1) it was

nccr:-ary to follow the disanpearare-c of the sulfur-to-netal

ci;a. iran'fr ak in Vt < njar-ultrav olet region. Here alvaan-

LagW could be L;cl oi the high (^l,b00j) molar abuorpLivity wiich

al'lo. -i so-ilI i' of extrr--l- 3o-r co cntration to ib used.








ABLE V

Stoichiormetries of the Reactions of
Cobalt(Ill) Complexes with Chromium(Il)


Co(ImI )L Cr(II) [H] Cr(III)L+ Cr(III)L2e Cr(Ill)L3
mmole mnoleo M role mmole nmmole


L = 2-Mercaptopropionate

0.100 .214
0.01CO .933
0.100 .141

L = Lactate


0.100
0.100
0.100
4.0 x 10-


0.241
1.00
0.245



0.251
0.244
0.249
0.050a



0.102
0.100
0.092



0.202
0.205
0.143


.0047?


.240
.232
.141
.0023


Methylthioacetate


0.450
1.00
0.150



0.500
0.250
0.150
0.050



0.175
0.094
0.065



0.400
0.208
0.064


.190
.190
.060


incomplete after 15 hours.


L =

0.100
0.100
0.100



0.100
0.100
0.100


L = Glycinate


-------


aReaction







In a typical experiment, solutions of cobalt(III) complex at

3.75 x 10 6 I and chronmiur(II) at 5.4 x 10- M each at H =
-2
1,0 x 10 -2 M and with ionic strength maintained at 0.10 1. witLh

LiCI 4 were deaerated while thennostatted,then transferred to the

drive syringes of the stocped-flow instrument as previously

described. Use of these concentrations allows the data obtained

to be treated as a pseudo-first-order reaction. Thus, a plot of

log (At A, ) time permitted calculation of an observed

rate constant nhich then could be converted to the second-order

ra r-"''nt, ''r -" cc.c-' tratiors spoified, 'the reaction

wvs co- plete in ?0 msec. To ascertain the dependence, or lack

thereof, on acid concentration and order of the reaction urith

respect to chro.iurz(II) ion, each was varied independently over

a ten-fold range while other variables remained constant. It was

found that the reaction is independent of the acid concentration

in the range [w+] = 0.10 i to [H+] = 0.010 1 and first-order

in chromium(II) ion in the range [Cr(II)] = 2.15 x 10-3 N to [Cr-

(II)] = 2.15 x 104 M (Table VI). By virtue of the linearity of the

log (.t A ) v' t plots over at least three half-lives (>904

reaction) it was concluded that the reaction is also first order

in o:<. int. The second-ordier rate constant for the reaction of

[Co(cn)2(00CCH(i .)S)]+ with chromium(II) was found to be (1.55

0.25) x 105 V-1 sec-L




c i a aas f found b ppro:i-

ri'-':' I 9 ijn":- ;n r ch: r.-C, n t at I ; = 0.010 (Table V) bv








TABLE VI

Acid and Chromiun(II) Dependencies of the Reactions of Some
Cobalt(IIl) Complexes with Chromiumn(II)



[Co(III)L]n [Cr(II)] [H+] kobs

M x 105 M x 103 I M-1 sec-I


L = 2-Mercaptopropioratea


1.05
1.05
2.15
0.54
0.21


L = Methylthlioacetateb


47.0
47.0
47.0
98.0
9.4


0, L.O
0.40
0,40u

0,40
0.40



13.6
* .o
13.6
13.6
13.6



48.0
48.0
48.0


aV = 0.100 ',


0.100
0.010
0.010
0.010
0.010


1.60
1.55
1.01
1.32
1.46


L = Glycinateb

8.40 0.100
8.40 0.010
98.0 0.100


0.100
0.000
0.010
0.100
0.100


2.22
2,22
2.26
2.05


I-~---~------


(LiC10O-4HCI04). b = 1.00 [ (LiC104--HC104).








rmethods analogous to those used in the preceding section. The

reaction was found to proceed by two different pathways, vide

infra, but usually only one chromium-containing product vwas isol-

able fro1 a given reaction. It has been found that the initial

chror.im -containing product of the reaction exists in two rapidly

interconvertible forrs, vide infra.

Py methods similar to those described for the mercaptide

analo-uo it was drtern ned that the oxidation-reduction reaction

was fiit order in cache reactant by pseudo-first-order techniques

(F ViT), It. r irsible to follow the disraearrance of the

proto. -: iorn orm f ', cobalt(ilI) complex by spcctrophoto etric

morn:to-rin-g it its absorption maxirmumi at 99 nm on the stopped-

flow ins truiient. Again linearity over >90, of the reaction was

observed for the kinetic plots. The second-order rate constant

for this complex iwas found, however, to vary inversely with

acidity over the range 1.0 M [H+] a 6.7 x 10-3 (Table VII).

A plot of kobs vs [H ]-1 mas linear (Figure 1) and yielded the

expression for the rate constant as


7.5 + 0.023 [H+1-1
k =+ 1 (if and sec)
1 + i- HL ]

Tlis observation can be understood from a rigorous solution

of the diffcrcntial rate equation for this reactionl0

-6 Co(T:i )t-.o' 3].
dt. = .OsL.obs o(1 )]toal (13)

i:,;;': s i.'i di ;oc -i ,.n constant of Co(en), ((;,.'C (CY:3)-
OH) 3-1 :*': k] I"-. k :rC defined by












TABLE VII

Acid and Chromium(II) Dependence of the Reduction of
[Co(en)2(OOCCH(CH3( )O1)]2+ by Chromium(II)



[Co(mll)] [Cr(II)] [H+] kob
1 i
x 10 l x 103 M M- sec


2.00 68.0 .0100 9.15

2.00 5.50 .0100 9.62
2.00 5.50 1.000 7.21

2.00 5.50 .1020 7,50

2.00 5.50 .0507 8.10
2.00 5.50 .0167 8.70

2.00 5.00 .0067 10.77

.10 1.00 8.00 x 10-5 47*


Appri;:-.iatc value; variations bet:.een separate reactions >20;.



































[14!]y i

of Qhc r. Ouctiori ofLOn2









[Co(en)2((OCCH(CH 3)OH)]2+ + Cr2+ products


-+ 2+ 1:2
) products



[Co(en)2(03CCi(CH )O)]0) + Cr --2 products

-2
From graphical analysis k2Ka is found to bt 2.3 x 10- and Ka

determined to be 4.4 x 10 vde supra, leading to the calcula-

tion of k2 = 52. Several experiments were performed at [I+] =

8.0 x 10-" M where the ratio of deprotonated to protonated forms

is -5 with the observed rate being 47. The large uncertainty in

rates (> <,') cr cc-'c:-: at this acidity lcvel, probably tLe' to

consu pticn of prcL.cns by the frec ethylencdianines released in

the reaction, renders the values approximate. Their usefulness

is limited to the observation that as the concentration of the

deprotonated form is increased the observed rate constant approach-

es the predicted value.

From the observed rate law it is obvious that at 0.100 14

acid the immediate product should arise almost exclusively via

the k, path. As the acid concentration decreases, however, the

product of the second path should appear and become dominant at

extremely low acid concentration. At 0.100 l acid [Cr(H20)5(00-

CCH(CH )(0)]2+ is the only isolable product. By using a large

volume and extremely low concentrations of reactants over a long

nriodc of ti.e, it is rnssihbF to observe both the +1 and +2

prodvc', in niol- r tio cf 2:1 at a p' =: 1.0 x 10-i ]'.

T:c ;'ace tLa;i pro'i ts are isolable front the reaction

is noL LaiiOu cienu t.o confirm Licir genesis by two separate

ci'ic path-'.;': i--Lnc b:,h, uro'u'ct; r ,re subject to -iubsequent







renations over the time period needed for the reaction and isola-

tion period. In fact, their rapid interconversion as a function

of acidity, vidi infra, over a time period much shorter than that

req'i red for isolation suggests a thermodynanic rather than a

kinetic distribution of products. Further consideration of the

mechanism is postponed until after the rates of substitution are

presented.

2- et."nylthioiacotaitobis (e thylenediamnin ) cobalt (III) Ion--[Co(en)2-



Usin,: i t ion e::c I C chrniques previcu-si detailed, ith

stoichLio.ctry of the reaction between chrc.iumr(I) and this ion was

found to be equimolar (Table V). The only isolable reaction priouct

was characterized as a +2 ion (Table I) and assigned the structure

[Cr(H,0)5(' *,*'- ) 12+ based on its spectrum (Table III). As
5 -1
will be discussed in the appropriate section, this species was con-

cluded to be a tertiary product of the initial oxidation-reduction

reaction, the result of ring closure and subsequent dechelation of

the primary rmonodentate product, [Cr(H20)5(CH3SCH2COOH)]2+

The rate of the electron transfer reaction between this co-

balt(III) complex and chromiun(II) was found to be conveniently

rcasurable on the stcped-i -flocw instrument. The decrease in

absorbance at 499 nm was followed spectrophotometrically under

rro-firr-"- co ," s al2loTin- data to be conveniently'

er .s.Q, .'' ;r n. A i.ical r ct/on vras un'r conditions

s' 5IT i< or Je sae ai Co follo-:ix : iCo(IlI) =1.3 x 10J

1i, [Cr(i )i = .'O x !0 Li I = O..I0O 1i and ionic strength
: oi







As for the previously described cobalt(ill) complexes, acid

concentration and chromium(II) concentration wcre varied inde-

pendently over at least a ten-fold range to evaluate the dependence

of the reaction on these two variables. Linearity of log (At- Am)

vs t plots over at least 90, reaction for the ranges [H+] = 0.100 m
tn o~loo ~pldC~r(;)II-2 -
to 0.0100 M and [Cr(II)] = 9.8 x 10-2 m to 9.4 x 10-3 I with no

significant deviation in observed rate constant leads to the

conclusion that the electron transfer reaction is first. order in

both oxidant and reduc .nt and inerepnendnt of the acidity withinn

the ranges specified. The calculated seconr-order rate constant

was found to 'e (267 18) -1 (a 1

Glycinatobis(ethylenediamine)cobalt(III) Ion--[Co(en)2(OOCC'C

.:i2) -

The stoichiometry of this reaction was determined by ion

exchange separation of the reaction products as described previous-

ly. The elution characteristics of the product are those of a +3

ion (Table I). Due to the longer reaction tine relative to the

three previously discussed complexes, the possibility of loss of the

primary product must be considered. Still, recoverable amounts of

the only isolable product, characterized as [Cr(H20)5(OOCCH2-

13,3)] +, v.~i sunra, approach closely those expected for equirlolar

stoichiometry over the range of reactant concentrations considered

(Table V).

The raie of th [Co(Rn),( .) 'lr(II) reacion was

found to t'L convenie.. s c u1 a ll- Ig-l sn aplroht'E

wrj.ti tne Cary 14 insil.rumnimL as aeucrljte in Lne rxperiinertal sec-

Lion. A ', i.cal recl' ion ,as rolutiior" I,; : 10al' : in tie


















^o "b "b



H H
H x C4

+1 +1 +1

0 N C\
vi 11;r


0 rH
+t +
+- +1



CO 0
rH


0 O
H-
+1 +1
'10 o
'"- -
CINf


O Hl H 0 0 0 0C O 0\ N0 C.- V 0 -

- C r Hr N c-


OC
0000
0 0 0 0
o C 0 rH
r-i r- 0


000
0 0 0
r-H H r- -
00 0


U'N V'\ 4 V't 0 0 0 0 0 0 0 0 0 0
0 0000000 000
0 0 0 0 } * * -1 V1
C . . Z C nd n .
H H H CO CO CO - - C c






0 0 0 0 N N '0 '0 '0 'C 0 0 0C
*.(r C(l' C11 (IN C( 0N 0 0 0d
S--. Hl rl r- u


C-'

U)

N


C)

(1
()
0


H-

U
,f U
0







cobalt co.i:plox and 8.4 x 10-3 M in Cr(II) at [H+] = 1.00 x 10-2

tj (ionic strength maintained at 1.00 1 with LiC104). These w-ere

transferred anaerobically to the glass mixing apparatus which ras

therost-tatted prior to reaction. Again pseudo-first-ordcr condi-

tions were maintained and the data treated appropriately.

In order to determine the acid dependence the [H+] was varied

from 0.300 M to 0.0100 .L with [Cr(II)], [Complex] and ionic strength

constant. The order of the reaction with respect to [Cr(II)] as

confiinX:d by use of the stopped-flow instrument with acid, complex

;r', iorTc 'rn th coles^nt and [Cr(II)! = 0.C0 x 102 M. Roe" lts

su ,ir zcd in Table Vi. In the specified ranges the reaction

was found to be first order in oxidant and reductant and independent

of the acidity. The second-order rate constant was determined to

bo (2.22 t 0.12) M-1 scc"1 (Table VIII).



Activation Parameters for the Reaction Between the
Cobalt(3Jl) Conolexes and Chromnium(II)



Activation parameters for the reaction of the cobalt(III)

cormnlxes and chromerin(TT) were obtained by determination of the

second-ordrwr rate constants for the individual reactions at temper-

atures varying from 13C to 450C, depending on the system. A plot

of log (knd order/T) vs 1/T was made as previously described. At

c .';t t. : .te'ai '. at ech of three dif 'rcnt tonroratures

-r" n:- 'c. t; r? ra.tr ,;:. rmainit in2. at 0.1 C by rethods pre-

vio-sly d-cribed. I'or the reactions at other than 25.0C, solu-

ti(rs s;er" ti
'1" *c-c o 1: nd vrl I.r t* t;' i l l or








at least one-half hour prior to reaction. Rate constants so

obtained for the four cobalt(III) complexes are summarized in

Table VIII.

Average values at each temperature were used to determine

the respective activation parameters via a least squares analy-

sis. Plots so obtained are shown in Figures 2, 3 and 4. Error

limits for the activation parameters were determined from the

most deviant lines still encompassed by all rate error limits.

Results are sur'arized in Table IX.

Attemnpt. to deter-ine activation parancters for reaction

involving th- deprotona 'd form of the Co(III) lactate complex

were abandoned due to a >20,o fluctuation in reproduciblity at

[H+] = 8.0 x 10- M where the deprotonated form is dominant.

This resulted in overlapping values of kobs at the various

temperatures. At intermediate pH levels distinct curvature in

the Eyring plots was observed. This curvature could arise from

different activation enthalpies for the two contributing paths

as well as from an expected variation in Ka with temperatures.

Thus, values at the lower pH values would not realistically

measure the activation energy for the deprotonated form without

a detennination of K and the acid dependence at each temperature
a
which was not done.



Su*bht i'-;-t " ,I : -




tc l c o' Co( )i,;jCC( ))t relative to

















2.9--


2.8-



I r - _




o 2Gr-
r ...-


2.5-


_______---II


3.10 3.15


3.20 5.25

10/T, (1K"1)


530


3.55


Figure 2. Eyring plot for [Co(en)2(OOCCH(CH3)S)-+.
I = 0.. 00 ( HC1 .,--1i.iC., ).


_~



































3.15 3.20 3.25 3.30 335 3.40 3.45
IO /T, (K-')


iFigure 3. Eyring plot for [Co(en)(CCHSC)]2+
Figure 3. Eyring plot for [Co(en)(OOCCHFSCH3))]2+
/X" .0 i (**-1' -LiClICO ).





































O/T, (K-')


Figure 4. Erin, plots for some cobalt(III) complexes.

-- .Co( .),(C01C-;( )
< := I.C [ ('C10--L "


















--I



S +1 +1 +1 +1



M -l r-I r-l r-l


















H C)
SN 0
+ +1 +1l






) u I CH C'r co
(SC. I I I






4+ 0 0
S- Er +1 +1 + 1 + I

r-1 0l r
SU I 0 z2 I C 0






r o
H 0 H0





o o2
(* a e +1 +1 +1 +1



0H H 4 ) a






QCCO)
I j
00h0


N 0I
+ -i* l
I .-N +
CL 0 I-- F--)











0 0 0 0 N
S C.'0 0
II

(i!








chromiiLm(II) at high acid concentration, the stopped-flow

instrument can be used to measure the rate of the first reaction

subsequent to the initial redox reaction. This secondary reaction

was followed by i-onitoring the increase in absorbance at 545 nm,

a naximun for the chelated 2-mercaptopropionate chromium(ll) com-

p3ex. This subsequent reaction has been postulated as the ring

closure reaction of [Cr (H20)5 (S(CH3)CCOOH)]2+ (mercaptide-bound),

the initial product of the redox reaction, vide suora.

The fi.it cxperir.ontally isolable product of the redox

rr'ctinr v"1-'- .!i cnoiticns has bch" ch arctnrized as the chelated

chro. a(n iG ) iroduc oI a this rinr closure, [Cr(H20)4 (OCCH(CC:i( )-

S)], ivide sruna. This product underwent a subsequent reaction over

a twenty-fcr-iour period which was then separated via ion exchange

to yield mostly +1 and +2 ions with eome small amount of a +3 ion

also present, especially a low acid concentration ( < 0.20 M). The

+1 ion exhibited identical spectral parameters to the original

chelated complex, [Cr(H20)(00CCH(CH3)S)]+, while the +2 fraction

was spectrally identical to the monodentate complex, [Cr(H20)5(00-

CCH(CH )SH)- The +3 icn was spectrally identified as [Cr(H20)-

6*. By spectrophotometric monitoring of the 264 nm absorption it
can be shoe;n that the +2 fraction converts back to the chelate form.

This peak has been assigned to a metal-coordinated nercaptide chro-

mophore on the basis of its presence in the chelate complex, ab-

.a';ic in tl4 oncdeLrt.e-t carboxyltc-bovnd co:nlcx and similarity

i5 cn.ra-: r7.- : ar alr orptivity to o'ter chro ::i (III )-nrrcaptide

Si- s (1%. 2 I). ii.cse obsrv:, iou. can be understood by con-

irTion ; i.c not ionic relations :l;.n observed rate







laws at constant acidity.


[Co(en)2(OOCCH(CH1 )S)]4 + Cr2+ + 5H+ -

[Cr(H20)5(SCIT(CHJ)COOH)]2+ + Co2 + 2en2+ (A)

I

I ~ [Cr(H20)4(OOCH(CH )S)] (B)

II


i + H+ [Cr(i20)5(OOCCH(CH3)H)]2, Keq (C)

IT!


d[II]
dt k[II] + k III], kb = k + k

The rate of the initial process (B) to form II varied

I;.r~o?,-):- r r;"i" ciiri.^ *'^: j. T~'7, of I v's [ I (Fi5_-re 5) rf'.e r"'7

is varied from 0.900 to 0.0900 H results in the expression k =

(1.0 + 2.6 [H+]) x 10-2 (M and sec, A = 1.00 i (HC104--LiC104),

250C).
A determination of the K for reaction (C) was necessary in
eq
order to cvaluate ki. ar kr, *.,r' snra Keq vas determined by

allowing a solution of II to equilibrate at known acidities. The

equilibration was followed sTectrophotometrically at 545 nm, a

maxiimm for II, over a twen1t-to thirty-hour period with readings

taken c-e t:''o '"-i for the first twenty hours. Since

tthe h i o- CI ,2),] -a .. overlaps ihe

c. ..... 1 on. a '' of A I3 time was then used to evaluate the

.o.... t q i.A. c wa~ic ciosc approxir a d by extra-
-, r :.' .. '* * ..i.... *:. asi n" [ r-ir al purtio ti o tf lirf




67
































0.25 O.O 0.75 1I.0
[Hi]



Finire 5. Acid dcncndence of ring clost]ur of
LCr(H2()) (SCH(CH 3)COO) I4 .







plot to t = 0 sec. This allowed a close approximation of K from
eq
the known nolar absorptivitier of II and III and the acidity. The

average value for three determinations gave K = 10.5 (Table X).
eq
The slow hydrolysis of III to [Cr(HO2)6]3+ (k = 10-" sec"l) gives

rise to a small inherent error in the Ke evaluations.

The evaluation of kf and k was accomplished starting with

solutions of II for the sake of convenience. The observed conver-

sion of III to II could have been used alternatively. The acidity

range examined was dictated by side reactions at higher pi to be

ex fh.rr '"uthc' .r CisccussncF else-ihere.

ire evaluation of k r t a given acidity was accomplished by
oos
plotting log (At- A q) vs t. The plots were found to be linear

over at least two half-lives. For reactions with [H+] < 0.30 i,

A was calculated from the value of K the acidity, initial con-
en eg
centration of II and the respective molar absorptivities of II and

III, Values of kf and k were then calculated from kobs, Ke and

[H+] by methods previously described.

Within the range [H+] = 0.65 M to 0.100 H, kf was found to

vary lin.arly with ::id concentration while kr was independent of

acid concentration (Table Ya). Plots of kf v s [H (Figure 6) and

k vs -r[ ~]l (Finnire 7) fave the following rate expressions:


k = (7.31 ['HI) x 10-5 (L and sec)

k = 7.10 x 10" (i and sec)

Errc r ]Air.i for t, rc4.pcct ivc plots of k ad k ;rC obtained

St ror i in i n c.luatio,. of kf and u .tbsed on
--i: c~?;.:li~i 83.:Co,













TABLE X

Evaluation of Equilibrium Constant for the
ChroniurC(III )-2-lMercaptopropionate Interconversiona


[HIc [Crl] [CrL]2+q K
-eq eq
S H. x 103 Mx 103


1.00 0.11 1.29 11.7

0.500 0.23 1.17 10.2

0.300 0.35 1.05 9.7

Avg = 10.5 1.2

a = 1.00 I- li:ii ,--LiC10O), 25C, 100 ml solution.












TABLE XI
Acid Dependence for Interconversion of the
Chromiunim(III) -2-Mercaptopropionatea


[H+] kobs k kr
5 -1 < -1 6 -1
1H x 10 soc- x 105 sec- x 10 sec-1


1.00 9.>' 8.51 .05 8.11 .67

0.650 5.85 5.12 .05 7.45 .60

0.500 4.26 3.57 t .05 6.83 .52

0.300 2.95 2.24 .05 7.12 .50
0.180 2.0( 1.34 .04 7.07 + .44

0.100 1.47 0.75 .04 7.16 .37

ap. = 1.00 (HC104--LiC104), 250C.



































[H"]


Figure 6. Acid dependence of ring opening of
rcra ^ ^T~cc" p43







































Figure 7. Acid independence of ring closure of
LCr(o ) (C: );:) .








When excess chromium(II) was used in the initial reaction of

chro:.iur.(I) with [Co(en)2(OOCCH(CH3)S)] no subsequent spectral

change indicative of the conversion of I to II was observable on

the stoppcd-flow instrument. Duplication of the conditions for

product studies yielded the chelated 2-mercaptopropionato-chrorniu.-

(III) complex in virtually stoichiometric amounts. These observa-

tions indicate that there is, in the presence of excess chromium(II),

another reaction of the mercaptide-bound pendant intenrediate.

This is must likely a second oxidation-reduction reaction in which

c --r-m(Tri) reacts .ith the .monodentate chromrurn(III) product

initially noried to produce chelated chromium(IlI) product and re-

generate chromium(II). In view of the latter product, this can be

considered as a chromium(II)-catalyzed chelate ring closure. Simi-

lar observation have been made for the analogous 2-mercaptoacetato

chroimium(lll) and maleato chromium(III) systems, respectively. '

Lactate as Lijand

The lactate ligand system was included in this study primarily

for comparative purposes in the redox relations with attention fo-

cused primarily on the sulfur analogue. As such, detail of investi-

gation into the chromium(III)-lactate system is less than that

previously described for the 2-mercaptopropionatc system, but the

experimental results obtained suggest that future exploration of

the oxygen system ereits consideration. The following observations

delineate ihe limits of our investi-ation.

As previously discus-c d, the monidentatc alkoxide-bound
iro:um(1.) special, [C)r(i2 )5(OC:i(Cii3)C006)j, has not been
ciiro~iiuiniitj~x upeect, tCr(zH2)0)u..(uCz(Ci3)Cuv)n)j has not, been







experimentally observable as a product of the Cr(II)-[Co(en)2(00-

CCi (CH3)O)] reaction at high pH and of the path inverse in acid at

low p~. The observable products, identified by chromium analysis,

ion exchange chromotoraphy and spectral studies as [Cr(H,0)5(OOCC--

H(CH )olj2+ and [Cr(HI20) CCH(CH )0)] arise in varying ratios

which are dependent on the acidity conditions. Further, the iso-

lated species are interconvertible as a function of the acidity.

Only the blue, +2 ion is produced under conditions compar-

able to the following: ['I+ = 0.100 1, [Co(III)] = 0.10 0.010 :

witih chrc4--.(IT) ion in excess, st-ichionetric or deficient amounts,

The ion-excr +2 ion, maintained throughout at pi = 1, yielded

the following spectral parameters: [A(e): 568(26.8), 413(33,2)].

Dropwise addition of 1.0 Ii NaCH to this solution to 2.8 = rp = 3.6

(via pH meter) produced a pink-orange species vhose visible spec-

trum had tho following characteristics: [A(6): 548(31 t 1.5,

438(38 + 1.6)]. Reacidification of the solution to pH = 1 regen-

erated the blue species originally obtained, viz., [A(E): 568(25

+ 1), 413(34 1I)]. If, instead of using an cluent of pH = 1, a

neutral eluent is used, the blue ion elutes as a +2 ion but turns

pink immediately on coming off the column. The pH of this solution

was found to be 2.8 and the visible spectrum was identical to that

previously characterized as the [Cr(H20)4(OOCCH(CH )0)], ion.

Acidification of this solution to pH = 1 regenerates the previously

c derihe | '(r("-0) ( C CDC:(e O) 'i)! srrecies as identified by its


0 t r i,
].i. i to i ou ,o:.:; i-::iibi y *that t<;i; con'rersions *,crc

si <]:'1, dru *;.-< r[ ot oi' t .r:r ,'er, .M ich sl onld be f.xtr -- rapid,







stopped-flow. kinetic runs were carried out. A solution of the

+3 ion, generated by conversion of the +2 ion through appropriate

adjustment of the pH to 3.0,was reacted separately with equal

volumes of 0.100 I1 and 0.200 I HC104 (v = 0.25 i (HClO4--LiCIlo)).

Concentrations after mixing were [Cr(III)] = 1.0 x 10- I, [H ] =

0.050 31 and 0.100 I_, respectively. Spectrophotometric monitoring

was at 438 nm, a maximum for the chelate species. A plot of

log(At- Am) vr t gave a first-order rate constant of kobs = 3.2 x
-2 -1
10 sec for both acidities, showing the overall rate process to

b,, -reu*r?.Tabl,. T Es, sc-,iC proton trnnsfcr appiar- excluded,

ihe lo-::.r linit of acidity used for the redox reaction (4.0

x 1.0 4) was dictated by the release of two moles of ethylenedia-

mine per mole of oxidant. Upon reaction the ph increases due to

the consumption of protons by the amnine functions thereby intro-

ducing the hazard of netal hydroxide precipitation. Using this

initial acidity with stoichiometric amounts of chromtium(II) ion

and cobalt(IIl) complex results, upon cation exchange separation

of the products using neutral eluent, in isolation of a +1 and a

+2 ion in the .olar ratio 2:1. The icrs were characterized spec-

trally as the [Cr(H20),,(OCCH(CH)O)1+ and [Cr(H20)5(OOCCH(CH3)-

OH) 12 species. Again variation in acidity produced interconver-

sion of the ions as previously observed.

Methylthiy oace-tate as Lir.and

The rl.i'lnt: hii bci i,:en this lircand and the nercaptoacetate

1;- 1, of wi;~.ch i iti, d civativ ar: the similarity of their

co ~ ll) ci'.plcxa. :viL c. a co;,ari on of t.ic behavior of the

cli-l 1di(J I.) iprduct'. i'l. ive to t previous results viti








i'nrcaptoacetate, any differences would be directly ascribable to

the transformation of the mercaptide function to a thioether funs-

tio-o.

The only isolable product of the initial oxidation-reduction

reaction in the acidity range 0.100 [Hr] 0.010 li and with

chro.ium(II) in excess, stoichiometric or deficient amounts was the
-2+
carbcxy3ate-bound chromium(III) product LCr(H20) (OOCCIHSCH1) I2+

the characterization of which has previously been described. The

ion underwent no reaction of interest other than hydrolysis to the
F-;.=; i[ ,f,- .; f ithr r i t4h ; hi ion cci.



By enploying techniques analogous to those used for the thio-

lactate complex, the author hoped to be able to discern formation

of a thioether-bound intermediate, thus confirming the bridging

lig and as i Ul tlioether raUier than carbonyl oxygen. Reactions

were performed using the all-glass mixing apparatus and the Cary 14

instrument. The reactions were monitored at 530 nm, an absorbance

maximum for the chromiun(III) chelate complex (by analogy with sim-

ilar compley.:s, Table III), and at 270 nm, the spectral region of

greatest difference in rolar absorptivities for the sulfur-bound

species rel .tiv to thFe c arhoylatc-bou nd menodentate chromiu(Illl)

product. The respective experimental conditions were as follows:

530 rin; rCo(TTT)-) = 5.0 x 10~-4 FCr(ll)1 = -.7 x l0"4 [IT+

0 : 2)(, C n; Co(T II) = .53 x

, ,; )] = 2.3 10 l i 1 0.1+00 i ad r -- 1.00 ii

(r. --LiaU) j,. in bo li cases a rapia decrease in absor-bwnce

,, ".....nr.' *o 0 %O rp:ctio-N n C fr t c" o:ir;'ion-rey1 uciol







was followed by a slow, small decrease in absorbance. The plot of

log (At- Ac) vs t was characteristic of two consecutive first-order

reactions subsequent to the initial oxidation-reduction. A plot of

b b
(1 1)Ao (1), + At
log A a vs time
AtA,,

for the initial rapid decrease in absorbance was characteristic of

a second-order reaction corresponding to reaction of [Co(en)2(OOCC-

H2SCH,)]2+ with Cr(II).

The subsequent consecutive first-order reactions can be .nder-

stood in terms of a two-step mechanism consisting of: (1) closure

of a first-formed sulfur-bound monodentate chromium(III) product

to yield the chelate [Cr(H20)4(OOCCH2SCH )]2+ followed by; (2)

opening of the chelate ring to yield the carboxylate-bound monoden-

tite [ 0 *' O()'C:'CL,. )]2+ opcies. The subseeqent first-o.ner

reactions observed can be rationalized only with great difficulty

if redox bridging is postulated to proceed via carbonyl oxygen.

Glycinate as Ligand

The inclusion of this ligand system in the present study was,

as in the lactate case, primarily for purposes of comparison in

interpreting reactivity patterns. As such, detailed investigative

work into the products of the reaction of chromium(II) with [Co(en)-

2(00CCH2H2 )2+_ was not carried to the extent of the 2-mercapto-
lp:cpioo:a ystLr:. As will bL discussed blow, limitations imros-c

by tic sie tc.; itcolf hinder complete work, but certain salient

features cf the chronirn(lll) rlycinate product were accessible.

In the range of acidity used (0.100 [ll ] 1 0.0100 i) thu

rctio o ) ( ) [ with crox-i'(I) or' in c-x,,







stoichior 'lic, or c if.cient amounts, the only isolable product

vas chn ezized as the ion [Cr(H2 0)5(:C'. 3)]3+ with spectral

par,-etc ; follo-S: [\X(): 573(22 1.1), 411(23.0 t 1.6)].

'ihe ion wrs fluted from a cation exchange column in the lithium

form with 1.0 I i The pV of the eluted solution was found

to be 3.5 + 0.1. addition of 1.0 MH iaOH solution to the

product solution ulntli pi! = 4.5 + 0.1, the solution color changed

immedia.tely front b.-o to green, but yieldea virtually the sane

visible specLrun as IGc original solution. After several days

*U-i;; c^!.r'1" '?' c"-'*"r ~'.r, a r!d-0-in ot a and seDsjat-i-o w:,s effeosted

using ion f:chanL c .qcs. A fraction presumed to contain a

+2 ion ,rs collected whichh exhibited the following visible spectrrim:

[A(6): 5j(33.0 4 5), 420(41 6)] (pH = 4.5). The large error

limits for the mola- extinction coefficients are a result of the

dilute solutions (10"' M) necessarily employed. From the rather

high values for the coefficients in comparison with monodentate

carboxylate-bound c romium(lll) species (Table II), the ion is pre-

sumed to bc [Cr(H20) (OOCCI 21ii )] 2+. The important features are

that a 1:1 rlycinn-chroniu:mi(III) product can be isolated from the

appropriate '. undergo a subsequent reaction to yield, in part

a chror )- .j ch elate cor lex.













DISCUSSION


The primary objective of this research was to better

define the influence which coordinated sulfur functions have on

the reactivity of metal complexes in oxidation-reduction reactions.

Coiclusions relating to this objective will be discussed first.



.- T i .. " ' I" "1



All redox reactions between the cobalt(Ill) complexes and

chro;ium.(II) were demonstrated to proceed by inner-sphere pathw.ys

through product analysis.

In attempting to understand the rate and activation energy

data to be presented, it is convenient to describe the nob process

for an inner-sphere electron transfer reaction as series of

steps o represented by the following equations (for clarj ty, only

the bridging ligand is represented):

k.
Co(II)X + Cr(II) Co(IlI)-X-Cr(II) K1 (14)
-1

k 2
Coc(T.I)-X-C-r(TT) [Co(TI)-X-Cr( T)' K, (15)



LCo(Ili)-X-Cr(J.) i Lco()-x-c.(rc) (16)
1-3








k-4



k5
[Co(II)-X.-Cr(TII)]* p- Co(II)-X-Cr(IJT) (17)



Co(II)-X-Cr(II) -- Co(Il) + X-Cr(III). (18)
K5
-5

Equation (1) represents the substitution equilibrium between 1he

reactants and bridred precursor complex which can rearrange to the

actjiated complex (2). Electron transfer is represented by qtra-

tion (3) while euazions (4) and (5) represent subsequent doaectL-

vation of the successor complex and decay to products, respective-

ly.

The rate of formation of the precursor complex can in certain

cases be rate determining. if the collision rate for the positively
9 -3 -1
charged complexes is taken to be 10 i sec the lifetimes of

thL rebulting outeGr-sphere encouaner coUUmplexes esmiated as 1 -
-12
10-1 sec, and the rate of exchange of a water which is coordinated

to chromium(II) and proximate to the bridging ligand is 109 1010

sec-1 a,b an estimate of 10 -1 sec-1 is obtained for kl.7C Should

k1 not be hde rate determining step, the stability of the precursor

complex becomes important as an equilibrium prior to the rate-deter-

mining step, formation of the activated complex. In this case the

free energy of activation can be expressed as AG = -RT In(K1K2).

This enables a discussion of the reactivities in terms of steps (1)

and (.) whether or not they are actually isolated in time.

This model provides a basis for the. discussion of the reacti-

vi- p-r re r., *ii;'" -< tu "'thr .l1; nrcv-o


eonprrii'on ('l"cbL~: .::;)


result." wA&c1


lvdv" <"- r rairp(















Vr' \1 -' ) V' ) 1 o o O 'd o 0 o n ') 4 bD b
H r l H l C) r- CO -4 ri -


0 N 0

r-1 r-1 1-1 1-


' 0 0 ON C'- .





c'-' O H 0' -( 1
C' Cg CO C- N0
(NI







C, r-4 C- 0
C H r-i
r-I


ri





< r-
0







I
ri






rF-i



r-l
H r0




rI
0
C)
) H




C l'



1.
0


















C)



C;


N N
I c


. *
CO N
I j


rH CO r .- ON NM
C C UO U\; c C N


S- CM (C0
C. H 4 \7 NCl 0


'.9 )"
0r C)lr


SC



*r *


43
4 ) N
4F-F

'0


+1 C)


I I I I I I I
N1

C, I I
vco~~o


4-'

H C) co CO N N
-I r- C- 0. 0 0'

r 4



O O CC
r- r1 -i -


4_ C)1 0 0 0
CO CCi


e-N
CO
r-4

ON


H1


0
Co 0 C ) 'o O3 -rl O
rH H rH *r *H


H
I-*(




0


0






O ri
:) -
4-4-1
g
EC F






C
-I C,



- 0







4,
ri



4'












01
fL;


S H l r 1 H H -H H


\0
'n, On \n

O
C' C> L 0
0







*r *r* .H rli


+ + + +C
I- --- ON r-- i




- ,- -- CON N- N M N C
N+ C CC C) + f cC' C M*
0 0 c 1 ) .) V Oz
I"- '-- : 0 '< H V N ) N' NCM + .4 .
-.) ; ) :U, ; U- 2;' L) CM C-M ;T CM 0. IT
C 0C C) C o I C ) C) O O 0 U NC 0 C 0 I
C 0 0 U U o U "C ) -1
^ L,' I m C ,u\ C1 - - u




C) O M O M C0 C C
F C I I 1 i) F ) I JJ i aJ I i J L J



U


1 (. f














,CX .ri rl o


o ) C)*-f)


I
r--



C!









rd
a








r. -1
o



4








H O
0
0
O) rC


C~) -' U.. ~ .


4-
m -CI 4
I I 00 C 0 0 0
I I N r-H C-\ H 0O H H
1CC



-







I I Cr ) r- o





H C) H 0

0 0 0 0 N ",) )O ,- )
N





Fr,, r- F-
o o o : o

0-CI O r-, r-. H r









S r








c o o 0 co -+- N+
-rl O O 0 6






rl *rI ri r




14 C + + +



0 N^ N CO UC N
0 - -- ---


I If O O C C C)'-'- -'
4} 1 0 C) 0
*, c i C -
Nt o IN N


H CO 0













4 O- 0 co
*cr















o
-P















S C) 0
0 nj 5

a) 0 C















r: ,C)
CD
c l I



) 0 00














c O
-P 0) 0

o *x






c .i
ci o .. l









I C O i

* *H o *,

, c r







*> Ij 1
TJ : 0 ,D
0 <
lUP 0r


r
u C

L'j



-3 i


j








For clarity the results will be discussed according to the sequence

of bridging functions (1) chclated carboxylate, (2) chelated

thioether (3) chelated thiolate and alkoyide. The order of

entries in Table XII is that in which they are encountered in this

discussion.



Chelated Carboxylate as a Bridging Function



All the complexes studied contained as potential bridging

i'u:.% coordi..t carboylate group in tidentae lands ith

the oihcr donor function also coordinated to the sase metal center.

This situation represents a departure from previous studies in

which simple monodentate carboxylate ligands were examined or in

whi ch a potentially chelatinf donor function remained pendant from

the carooxylated metal, It was, therefore, deemed essential to

establish any distinctions in bridging efficiency between the

carboxylate group of a chelate and those previously studied. For

this reason the reaction of [Co(en)2(OOCCH2H2)] 2+ with chromiumn()

was investigated.

Entries 1-7 of Table XII suramarize prior experimental results

for the types o' carboxylate coordination previously studied. The

acetato complex, entry 2, can be taken as the prototype. The more

r"'~' r-te of reduction c~l~cr"nd for ti" fnrmrato ce-nlex, entry 1,

rrl.: ; in t!,- er' "1. co-1: and carn i ttributri to a diin-r

i:: ;ric 'r :estction : ;ct: is e':~pciitd to L: concentrated in

a Greater stability of the precursor complex, KI. Vihile the rate

of' ,'" re -'ic:rl'y ri. <-.1" io-b, '- -:+o cc-ip1. ent ry 3. is








dilini::r.d as expected, the source of this djrinuticn is found

1i', '1.r in the enthalpy of activation, a result not understood

by ths author. The diminished rate for the pendant glycinato

corni:e, entry 44, is attributable to the entropy difference, a

logical consequence of the increased charce.15 Entries 5, 6 and 7

are for cormlexes with pendant functions which can chelate the

chronmiu(l) redr tant. The expected greater stability of the

preclrcor couple and enhanced rates are reflected in more favor-

able e: 'ropy cor.tri itions.

in this coo'iter t.he enhanced rate for tl:e chelated 'lycina+.,

co00l;. ntr l**ears easily understood. 2he acceleration

finds its source in the entropy term. This is ascribed to a

gre.t:' slabilit; of the precursor complex h.en the carbonyl func-

tion to which chro'dum(IIn ) rost probably birds5 is held in a nore

accessible position as a result of chelation by the amine function.

This effect is expected to extend, with allowances for variations

in charge type, to other chelated carboy late ligands, thereby

fulfilling one objective of this research.



Ci i,] t I_!'ljj o -' r -- I r.'. n' uIrCt' f'-'ri



For the important case of thioether coordination in [Co(en)2-

(OOCCriS CH )] entry 9, it was not possible to define the bridging

fr-nctr throw ih irjolation of the chrorn"-(III) product which was

ala)r')-, "..^ to C'-> Cr(.C) 5(rCC '*'hilc this result

... >;.i :. i rc 1 rid ;i. I P(~ il rTi g i to con-

riti ; l , c" (> -: itl r: i, I' 1fur i 1-







[Cr(&20) (' .4 200H) + Such a consideration is mandatory in

view of a rate substa-tially greater than that found as typical for

carboy-late bridging even in a chelated example of the same ehar-ge
ty 5
type, vidp surra. Further, previous research suggests that

chelate closure of the alternative product to yield [Cr(H2O0) -

(I, SC'2 CO0)]2 could occur in times shorter than those required

for isolation at the acidity level (0.100 1') of our experiments.

A relatively rapid hydrolysis of the chromiriu-thioether bond, which

would rnot be surprising, would lead to our product observations.

In this context it should be noted that the relatively rapid

redox rate observe'; arizcsC exclusively front an entropy contribution

which is-12 eu more favorable than for any previously studied

carboxvlato-bridged reaction lacking a pendant donor function.

Further, the absorbency changes at 530 nm and 270 nm reveal a

sequence o' tnree steps whicn cannot be ascriceo. to carooxylate

bridging, a mechanism which should result in a single-step absor-

bency change. However, the two substitutional processes previously

described for [Cr(H20) (CH3SCH2COOH)3 could account for two

absorbency changes subsequent to the redox step. These results

are taken as indicative of-bridging via the thioether function.

If this interpretation is correct, the inner-sphere reactivity

bestowed by a coordinated thioether is one of very few examples

lying intermediate between that bestowed by very efficient bridg-

in- lirnrk., e.g. the halides and thiolates. and by the rather

rodcerr b~ritd!in" lil-nids, e.g. ilt.er and corb xylate. Thus, a

;,. obj-c or thi st iy seer's '' 1if l d.








A detailed discussion of the reactions of the [Co(en)2-

(CH SC, 2C~,212) 3+ complex, entry 10, is appropriately deferred to
46
another thesis. However, its greater diminished reactivity is

attributable to a substantial decrease in the entropy term while

the enthalpy actually contributes in the opposite direction. In

fact, its reactivity parameters, in comparison with those for the

complex described here, are decidedly at variance with those anti-

cipated for an inner-sphere reaction in which the non-bridging

function cis to the bridi-in ligand has been changed from carbox-

'Jr4",- t-' c.in-. fo cQ *-,7 rer I" in litti." arition 'n a-.

paro --tcrs for ori cori sjrinon, entry 11 -vs entri 1, and little

variation in the observed rate for another, entry 12 vs entry 13.

A sonevi;hat different pattern emergCs for two other inner-sphere

reactions entailing a similar variation of the cis non-bridging

function, entries 14 and 15. The reasons for this different

pattern are not yet understood. However, if the more rapid rate of

reduction for the [Co(en)2(OCCH 2S)]2+ complex is ascribed to en-

thalpic and entropic variations similar to those used for the thio-

ether comparison a value for the entropy of activation is obtained

which seems unrealistically high. Thus, a different comparative

pattern front those previously observed for consistently inner-sphere

reactions is evident. The results reported here may prove useful

in assiFning the reduction of rCo(or)n(C-'.. '-.--. '"-)13 to the

oinrco,' !re c''-ol. '-. "ho aotvatU-p'i prxrarseoro ore in sul7stntji-

al[ i r -.ont Co..' o e ..f r c ;..[;'r-;'.'^hc.r'-: r- ctio ; ir +3 to, ,


leIT C i : e.








T i '. I "r. i i'.'U .- _-_ :,-. i' r I .: :U



In the pioneering research in this area, Lane found a

reactivity for coordinated thiolate as a bridging function which

exceeded that for a comparably coordinated alkoxide by a factor

of >3,000.1 Three possible reasons for this enhanced reactivity

were presented: (1) a greater stability of the precursor complex

with the thiolate ligand arising from the greater steric accessi-

bility of the sulfur atom, (2) a cobalt-sulfur bond which is

wor.e.!r tin the coialt-cnxygcn bord,thercby requiring less entha -y

for activation of the precursor complex, (3) a possibly greater

sigma covalency in the cobalt-sulfur bond which might contribute

to an enhanced probability for electron transfer. io distinction

was possible between the relative contributions to the reactivity

from these sources.

The initial phases of this study were directed toward

providing such a distinction. Space-filling models suggested that

the methylene hydrogens on the carbon atom adjacent to the coordin-

ated chalcogenide would inhibit precursor complex formation with

[Cr(H20)5 2+ to a greater extent for alkoxide than for the larger

thiolate sulfur. According to this view, it was felt that substi-

tution of one or two methyl groups on this carbon atom would dimin-

ish the rate of reduction for both comulexes via a steric effect

without drastically alterin:- the electronic contributions. (Unfor-

tunately attempts to prepare tih dimrethl derivative I.ere unsucce:s-

ful, bat die monomunhyl derivative proved to be synhlotically access-

ible.) It was naively assuren that the reactivity of Lhe thiolate








coolex. vould be Iless sensitive to this change than the complex

w'ith sm.ll.er, 1-ss accessible oxygen-bridging atom. Finally,

if th< txrecCled ci inutic- s in rates actually materialized it was

hoped UI't a ccteri;ination and comparison of the activation para-

metercr. vold be 1 :.

The antici ated decrease in the rate of reduction was, in

fact, ( servedd as a conmarison of entries 12 and 21, 14 and 22

indie s. .singly, t: e factor by which the rate is decreased

is -"/' 0 for bo~ the alko .de and thiolate complexes. This im-

pli"s t 't any rreatcr antic: crated storic susceptibility to inhibi-

tior. the a ':'-d' frl-.ti:e compared to that of the thiolate is

not de ue.oped ly : r,.. r', substitution. For both complexes the

doer: -: in the r.cccssirlit of the bridging atom would, in the

abse-cc of sufficient activation parameter data. appear to be

compAr:..oJe Irmpyini a nignly carectional approach 2or the LCr(iQ,)-

5]T residue (i.e., the methyl function exerts a restrictive influ-

ence but can be comparably avoided in both cases).

In the ca;e of the rercaptopropionate complex it was possible,

as t-r 1 ut-icipatnl result of the decrease in rate, to measure the

actiwhtion par yrleers. The enthalpy of activation of 1.1 kcal/nole

reflec'i an un ially sc 1al resistance to reaction from this factor.

The r cson for ic differ ce in this parameter in the reduction of

the i:"rcaptoet}t l ,ine exn ple, the only other case for which it

har }ic-r <(ctr is not presently understood. The latter con-

l" '.. a .o v'is wt c-"'rd to the enthalpy contribution.

'. : "' ioi 'r i visible ren su:ests

sc- ': Ix :. lc:onc Tration. Even if this







were not the case certain differences should be recognized in the

two ligands. 'The mcrcaptopropionate ligand possesses an sp carbon

in the chelate skeleton whereas the mcrcptoethylanine has only

sp carbon atoms. This difference could result in different con-

formations for the two chelated ligands. The consequences for the

activation parameters of such variations in chelated ligands is

essentially unexplored. Further speculation is best postponed un-

til nore data are available.

The observed entropy of activation for the mercaptopropionate

corr',lx seems remarkably positive for a soecics with a methyl group

and a. hydro ;:n atom on a carbon aton bound to tie bridging atom and

further constricted by the chelation of the ligand. In fact, the

steric restrictions appear to leave the sulfur as accessible as the

halide-bridging ligand in cis-_Co(en)2(Cl)(H20) I+ and cis-[Co(en)2-

(')(Hi0)I2" and, in spite of the methyl substitution, less restrict-

ed than the alkoxide oxygen of the unsubstituted glycollate complex,

vide infra. These entropy trends are taken as indicative of a

uniqueness of the large coordinated sulfur atom in remaining

sterically accessible in spite of rather bulky substitutions.

The activation parameters reported in Table XII for the gly-

collate complex, entry 12, are to be regarded as tentative and

subject to confirmation. Nevertheless, they appear reasonable in

that AH lies intermediate between that for a nercaptide bridge

in a ccnara.ble environment, entry 22 (ccmncar also entries 23

pr2) ?'), ard t'-at repc,"ted for [Co(T: )- O )'"' ,-wich ChMnld have

a ha r A6' the r- rIt of the ,gne in no-iridin; lif;Lnrc;

c.o nr entries 1 and 11. The val]c for ASC a'pp'rn:- te.








The di cr'c~ ncy bot -r: n the tentative value for AE of 2.2 kcal/

role :;iu the 5.1 kcal iole observed for the reduction of Co(eon)2-

(i' )12+ | a'y arise from an enthalpic anonaly for the latter

similar to that s a acted for the [Co(en) i'.': -' )+ analogue.

If theL are vicd s anomalous activation enthalpies it seems

possible, to obtain tentative estimates for four unknown activation

parade nhich ppiar reasonable, internally consistent ana in

satisfcc ,y cos pi.on in the entropy term with the measured val-

ues foL aiion.lous complexes.

'"- tir-' rocc follows. Since the effect oC

; '**'> o on the a Ojacent carbon is primary :teric and is

com p .. for boti alkoride and thiolate complexes, it seers rea-

sonabltc t attr1.-l the rate of decrease to the entropy term. If

the val.. of 1.1 kc1/mole for AH in the riercaptopropionate reac-

tion i, uscod for the ricrcaptoacetate reaction a value for AS' of

-25 ei, is obtained, The increase from -31 eu seems reasonable for

the ]o:;s of the ". yl substituent. Further, the increase from

-.6 eu 'or the o:'. .n analogue is comparable to the 14.7 eu in-

crease r'e rved ror a similar change in going from [Co(en)2(OCH2-

};.:" ) to Co(c n)(-.': )2+ Proceeding in the reverse

direc:" A o2 kcal/mole for the glycollate complex is

assun r for the Jc .Le complex. This yields a value of -43 eu

for A' h'ci i 'in -".2 eu more negative than for the sulfur

.~--t ' n ".ivi than fc reaction :with the co -pl]x



S. Lt:s i, -i c; Ir e onblu Laii follcn -

i !, ":i !"**," 1~:.1' in!fi- nrC- reaction can be







2
dranm. (1) The carboxylato-chalcogenide chelates with an sp

carbon ator in the i'vc-membered chelate ring bestow an entropic

barrier to activation whichh is about 8-10 eu more negative than

2
for the anine-chalcorcnide chelates with no sp carbon. This

difference seers reasonable in view of possible confonmational

differences mentioned earlier. (2) The substitution of a methyl

function on the carbon adjacent to the chalcogenide atom in the

earboxylate-.chalcegenide chelate increases the entropic barrier

to activation by 6-7 eu. (3) The substitution of sulfur for oxy-

cn n n ot' r-es ar.alo-o's alkoxide complexes lo'.'.rs the entroaic

:barier 12-.5 eu ua drile the enthalpy decrease contributes about

one order of magnitude (1.4 kcal/mole) to the reactivity. Thus,

at. east for the carLoxylatc-chalcogenliue ligands, the enhanced

reactivity on substituting sulfur for oxygen seems to derive about

35i, from a lowering of AHr and about 65, from a nore positive AS*.

iherofore, the steric component associated with the larger sulfur

atom appears to be larger than the electronic contribution to the

observed enhancements. Thus, within the limitations expressed

earlier, a third objective of the research seems reasonably ful-

filled. In this regard we wish to acknowledge our indebtedness

to the research of Rolert !H. Lane and ]iichael J. Gilroy without

which the necessary corinarisons would not have been available.









As pr1:;i'lyc ur.-cFrl-'d, the :;uW'; titei on lie ':avior of t;,







2.-mrrcaptopropionate-chrori4umn(III) product ray b, represented by

the following steps:

[Cr(H20) (SCHI(CH)COOi)] 2+ k [Cr(li0)4(SCI(C 1)COOH)]+

I II

k
LCr( O)4(SC (CH3)CO0)SIT+ H
[cr(2o0) (SCH(CH3 )00o + H [Cr(H20)5(OOCCH(CH3)Si)-
+


k = (1.0 + 2.6 [H+]) x 10-2 (M and sec)

kf (7.31 [HIJ) x 10-5 (IU and sec)

kr ='!'10 y !0-6 (r, a"I v-e:c)

These are to be compared to the analogous k, kf and kr for mercap-
10 48
toacetate0 and for malonate:

Hercaptoacetate as liyand; -'alonate as ligand;

3 *) x K-3 (4. + f) x IC-6

k = (73 [H+1] 0.6) x 10-6
kr (68 + 0.8 [H+]-1) x 10-7

The initial chelate ring closure will be considered first.

In the first report of an analogous chelate ring closure by car-

boxylae whici follows a rate law of this form, the proposition

was advanced that the acid-dependent ring closure involved substi-

tution of a coordinated water at the carbon of a protonated carbon-

y3 function. 'lhe possibility of a similar substitution by coordi-
nated roxid i in the ac)o-inacpendent path was recognized but,
ld rs r r!j- iy not stronr-lv advocated in view of the alternative

S. ;;. cr : c Si : conc i lons vere
r'-"'" irt 10 C;,.t:O of' I." ml'crcapt+ ":**i't'e weiCr'' tnh- Yl,000-fold







enhancement in rate vas primarily attributed to the higher steric

feasibility for closure of the five-mombered chelate ring than for

the six-membcred ring of the malonate system. Sone contribution

arising from an enhanced nucleophilicity of water or hydroxide cis

to the mercaptide donor function was also reco:,nized.10 These

schemes are considered to apply to the 2-mercaptopropionate system

and are outlined in the following diagrams:

First-or'der acid _ath Acid-independent path


[(H20) 5c:(sH(cH3)COOH:)] 2
+

11+
lI[K


(H20),Cr' C -CHi
2 \ \ 3



113

(H120) Cr O. C'
+2+ .- CH2
+211 + H120


[(H 20) 5Cr(SCH(C3 )COOH)]2+



1H 2+-
/ "-C--CH,
(H20) Cr \ O-
H OH

+
H C
/s, 0"1i
(H 20)4Cr \
0


+H+ + H20


KH = 2.6 x 10-2 koKo = 1.0 x 10-2

The essentially unchanged ratio for the two paths at constant

acidity in all three systems lends some credence to the proposed

i .;' -.:*:. r ',, r" ;!.a::i: :: :,; : c 1- to :- ub ti1::-i.:C l ;? at c! ro h,;lI in

S r, t r rLthcr, tin, sii l;'.rt cr:iancc. :ntobserwd

i n'r tl,; r- c:rc pfto(p io'::ate s,,;; Ir in boil, :aths o-'.'r ,'e i;cr-

c't.)oeetiato srystctm is that anticipated for a sv "htlyv r"hanc:ed

- 1n '- ,; ": a I- I :ro ") ,,*, ,-, i ,on.








A comparison of thu data for the ring opening and closure

of the chclate co plex a the :retal-silfur bond provides the most

striking dcfercnce bet. -n the two r reaptide ligands. For the

methyl-substitu ?d liga- no acid-irndependent tern for kf (acia-

dependent for k ) was obsr-red whereas the mercaptoacetate system

has a term. A discussion of possible reasons for this will be pre-

sented later. Tie acid-dependent term for kf can be rationalized

by the following mechanism:

+ 2+

2 0 --k C
(t^2)ur.^o / 3 + T (n20) 2__ -3 k-
C. 0-C C r


[(H2 0)Cr(OOCOC((@ )SH) 2+

k = Kk


Such a mechani.l:i, has beon1 proposed for the analogous path in the

mercaptoacetate case and from the similarity in the rate laws is

similarly reasonable here. On the basis of inductive effect, it is

somewhat surprising that k is not larger due to increased basicity

of hlic mercapi id1. IHo xui, a sligl:.tly larger K right be compen-

sased by a co,- orably ; 'ller k. Any variation is expected to be

sr -1 in vi I o the f h t t the equilibrium constant for the

dechelation process (105, .12) is experimentally indistinpuish-

able from tiht. y pan nri s rocr, Winith nrrcatoacetate as li-and


i ; i occsa to t !' o-

[ox a 1 ,.. u ( .1 x 10 vs .b x -U )







Thus the arguments presented for that system are applicable here

and will not be repeated in detail. Thus it seems highly probable

that bond-naking is more imporLant than bond-breaking for both

forward and reverse processes. No evidence exists to suggest

that the acid-catalyzed metal-sulfur cleavage differs dramatically

from the analogous metal-fluoride cleavage in [Cr(H20)5F]

(k = (1.38 x 10 ) [H1+])9a except in the greater basicity of the

coordinated mercaptide. Further, the rate enhancement for chelate

closure by the mercaptan function by factors of 30-250 over the

rates for monor'ntate liration of [Cr( -'0), ] by PF(k = 2.2 x .10
8 49
of E3 (k = 2.8 x 10 )9 appears primarily attributable to a

"chelate effect" in which the pendant mercaptan function can better

trap a vacated coordination site rather than to a higher associa-

tive reactivity.

With regard to the lack of an acid-independent term in kf

(acid-dependent for k ) for the mercaptopropionate system in

comparison to its presence in the mercaptoacetate system, it should

be noted that by necessity our experiments were carried out over a

higher range of acidity (0.10 M 1.0 1) than that for the morcap-

toacetate case (0.010 M 0.20 M). Thus, within probable experi-

mental error, the zero intercept for the kf vs [ i+] plot is likely

to be indistinguishable from the anticipated 0.6 x 10- previously

detected. Further, a line of positive slope drawn through the

error limits of o.r most reliable exp-r.nents (<0.5 Ei iin the k v

[1!l plot wo-ld allow for a 0.1 x i0-0 [!+] contriu-ition to kr

while. rcq:iri; oonly a : 1 revision to 6.1 x 10" in the acid-

in,] '"rdent t ,n. In vi'- o' tho 1s critics o .he lirds: it







seems likely that this path would disappear and more reasonable

that, under the conditions accessible to us, it is undetectable.

The fact that the acidity range previously studied with a

very similar ligand-complex system was not accessible with 2-mer-

captopropionate is of obvious interest. Equilibration experiments

below 0.10 M acid were irreproducible and the odor of H2S was

clearly detectable. A previous study of cis-[Cc(en)2(0)HW!2CH2-

CH2Br)Y+ revealed a lysis of the nethylene-bromine bond by attack

of coordinated hydroxide on the saturated carbon atom to produce

[Co(cn)2NH2CH2CH20)2+ and Br.37 Thus, the most obvious explara-

tion of our results invokes a comparable lysis of the carbon-sulfur

bond.


a 7+ K+ -
[Cr(H2gO0)5L(OOCCH(C3 )Sh) == cis-(H2 ii20Vur" -- C
H SH

Ikl


,0--C"-
(H20)+Cr,O--- H + H2S
H I
CH 3

Consideration of the S2 mechanism for ihe attack of coordi-

nated hydroxide on the saturated carbon suggests that for both

steric and electronic reasons the presence of the methyl group on

the carbon atom of interest should decrease the tendency toi:ruds

direct attack.50 Alternate mechanisms, however, can be defended

only with difficulty relative to the observed product behavior.

Until further planned work on this system is accomplished no defin-

ite conclusion as to the reason for the reactivity of the complex

can be reached.








Chromiun(III)-Lactate System

For the lactato and glycollato complexes of chlromium(III),

the lack of extensive data precludes a discussion in the detail

afforded the mercaptide systems. For the reaction
k
[Cr(H20)4(OOCCH(R)O)]1+ k [Cr(H20) (OOCCH(R)OH')]12+
r
the values observed for kf at 0.3.00 11 HCO4 and 250C were very

nearly identical (R = H, kf + 3.0 x 10-2 ii- sec-1;l0 = CH13,
-2 -1
k, = 3.2 x 10-2 -1 sec1 ). In view of the similarity between
10
the two systems, the tentative conclusions reached previously

are applicable here. It seems likely that the mechanisms parallel

that presented for the analogous nercaptopropionate reaction. The

higher rate of ring opening is reasonably attributable to the higher

basicity of coordinated alkoxide vs mercaptide.

It seems appropriate to recognize the similarity in spectral

parameters for the chelated lactate and glycollate complexes in view

of the discrepancy Twith earlier work mentioned by Lane.10 These

results with a very similar ligand provide further evidence of some
39
error in the earlier report.3

Other Chrc-iud(IIT) Systems

Similar substitution studies of the chromium(III) complexes

with methylthioacetate and glycine were of secondary priority and

work other than the identification of the reaction products of the

initial oxidation-reduction imas not attempted. It was noticed that,

upon appropriate adjustment of the acidity to lower levels, [Cr-

(H20)5(OOCCH2Ni3)]3+ reacts to yield an icolable product tenta-

tively identified using spectral criteria Ls the chelate complex,

vide pra.








S iIax -


This research has led to the following conclusions.

(1) The incorporation of the carboxyl;rite function in an appropriate

chelate ligand can increase its bridging efficiency, apparently

by increasing the steric accessibility of the carbonyl oxygen.

(2) Cobalt(III) cormplexes in which a coordinated thioether function

is the tenninal donor in a chelate ligand can be prepared (conclu-

sior reached jointly with Michael Gilroy). (3) A coordinated tnio-

tirr frc~retion r'n serve as a bridging liBand in an inner-spherr

reaction r:ith an efficiency which is inter~ diate between hiuilj

efficient and poor bridging ligands, (4) The chronium(III)-thio-

ether lond in the secondary chelate product of this reaction ie

quite labile. (5) The substitution of a methyl group for a hydro-

gen on the carbon adjacent to coordinated chalcogenide does inhibit

inner-sphere reactivity but, surprisingly, by a similar factor

(-1/40) for both oxygen and sulfur. (6) As a result of the inhi-

bition, the activation parameters for the mercaptopropionate sys-

ten proved accessible and suggested from the high entropic barrier

to reaction that the inhibition was steric in nature as anticipated.

A cornarison of the p.careterzs with rcsilts obtained in this labor-

atory by co-workers permits a reasonable and internally consistent

estimation of activation parameters for other reactions. Analyses

of t.' -"atio: in t -se 'Tr.c-ters prir-,iL a rn.ch nore detail: ]



; o C: I t l t a Ii iand thi 'as

puon"."Jc -iri i< "c 'or t l <:id cLct ;osi tiO:. t' it'L
pc< ::'. ,]c i. <"`vio ;r o t} <. ix!



99


the high reactivity in derived mostly from the steric accessi-

bility of the large sulfur atom. The lower reactivity of coor-

dinated alkoxide complexes arises primarily from a lorwr accessi-

bility of the smaller oxygen atom. ('/) The presence of the ad.a-

cent methyl group apparently enhances the susceptibility of the

pendant carbon-sulfur bond to lysis by coordinated hydroxide in

the chrcmium(IIl) product. (8) Finally, we note the preparation

of a number of new complexes via oxidation-reduction reactions.

Thus, the objectives which were outlined for this research in the

Introd.stic soen fj.fille"d













LIST OF REFERENCES


1. (a) A. San Pietro, "Biolo(-cal Oxidation," T. P. Singer, Ed,,
Interscience Publishers, rwr York, 1968, p. 515.
(b) K. B. Rajaropolan and P. Handler, ibid., p. 301.
(c) J. C. M. Tsibris and R. W. Woody, Coord. Chem. Rv., 5,
437(1970).
(d) 0. Hayaishi and M. Nozaki, Science, 164, 389(1969).

113(1967).
(f) R. Kimura, t. n,-..... and Bonding 1(1968).
(g) J. C. Rabinowitz, Adv. C .. Ser., 100, 322(1971).

2. (a) J. R. Herriot, L. C. Sicker and L. H. jenscn, J. Mol. Bil.,
50, 391(1970).
(b) J. C. M. Tsibris, R. L. Tsai, I. C. POnsalus. H. H. Orr."-
Johnson, R. r. nansen anli n. neiner6, }iroc. i.atl. ;cad. i..,
U. s., 2, 959(1968).

3. (a) W. A. Eaton and W. Lobenberg, J. Amer. Chem. Soc., 922,
7195(1970).
(b) W. Lovenborg, private conmunication.

4. R. E. Dickerson, T. Takano, D. Eisenbcrg, 0. B. Kallai, L.
Samson, A. Cooper and E. i argdiash, J. Liol. Chem., 246, 1511
(1p71).

5. (a) A. San Pietro, Ed., "Eon-heme Iron Proteins: Role in
cr'-y Converion, Anticc Press, Yellow Springs, Ohio, 19.5.
(b) Poe, V?. D. Phillips, C. C. ,IcDor.ald and '.. Lovenberg,
Inc. 1atl. XAc1?. Sci., U. S., 6 7"?(1970).
(r) D. h illins V. o C. C. K'cbonald and R. G. Bartscl,



C -'
7. &) .i E. i 'orl .i :on, C. ', lenti ad E. an,
".o iil :._ .- 1, 1 Tsf ris,







G. (a) M. Schubert, J. Amer. Chem. Soc., 54, 4077(1932).
(b) D. L. Leussing and I. 11. KoltIhoff, ibid,, 7. 3904(1953).
(c) 1. Tanaka, I. M. kolthoff and W. Stricks, ibi.,, 7Z,
1996(1955).
(d) D. L. Leussing and L. Newman, ibid,, 8, 552(3-956).
(c) H. Lanfrom and S. 0. Nielson, ibid,,Z2, 19 6(1957).
(f) J. E. Taylor, J. F. Yan and J. Wang, ibid,, S, 1663(1966).

9. R. H. Lane and L. E. Bennett, ibid, L2, 1089(1970).

10. R H. Lane, Ph. D. Thesis, University of Florida, 1971.

11. (a) H. Taube, Advan. Inorf. Chem. Radiochem. 1 1(1959).
(b) :1. Taube, Can. J. Chnn., 27, 129(159).

12. V., r'n'.b r r. "Y ers and R. Rich, J. Aror, Chem. Soc., 7,
4118(1953).

13. H. Ta-be and H. iyers, ibid,, 76, 2103(1954).

14. (a) H. J. Price and H. Taube, Inorg. Chem., 7, 1(1968).
(b) J. F. Endicott and H. Taube, J. Amer. Chem. Soc., 86,
3653(1964).
(cr' i. J'aI3, 0. Cb',". P.. 45, 45r2(1.8).

3.5, E. S. Gould and H. Taube, Accts. Chem. Res., 2, 321(1969).

16. I. A. Alexander and D. H. Busch, Incrg. Chem., 5, 1590(1966).

17. J. 1. Worrell and T. A. Jackman, J. Amer. Chem. Soc., 93,
10l'4(1971).

18. F. S. Gould, ibid,, 8, 2983(1966).

~9. 1. J. Gilroy, work in progress.

20. H. 1litz and ',. Blitz, "Laboratory Methods of Inorganic Chem-
istry," John Wiley and Sons, New York, N. Y., 1909, p. 173.

21. Linhard and !H. Weigel, Z. Anorg. Che ., 264, 49(1951).







25,. t f V .Jc h >- :'. c'b, ('. e. (I, 24 153 ?.)
24 ,. . ..







26. T. J. Meyer, Ed., "Gmelins Handbuch der Anorganischen Chemie,"
58A, Verlag Chemie, G. M. B. H., Berlin, 1932, p. 497.

27. H. Sueda, Bull. Chem. Soc. Japn, 12, 483(1937).

28. M. L. Moss and M. G. Mellon, Anal. Chem., 15, 74(1937).

29. D. A. Schooley, unpublished results.

30. M. V. Olson and H. Taube, Inorg. Chem., 2, 2072(1970).

31. A. A. Frost and R. G. Pearson, "Kinetics and Mechanism," 2nd
Ed., John Wiley and Sons, New York, N. Y., 1961, pp. 172-174.

32. A. A. Frost and R. G. Pearson, ibid., pp. 95-96.

33. K. Nakamoto, "Infrared Spectra of Inorganic and Coordination
Compounds'; 2nd Ed., Wiley-Interscience, New York, N. Y., 1970.

34. M. Ardon and H. Taube, J. Amer. Chem. Soo., 82, 3661(1967).

35. R. Butler and H. Taube, ibid., 87, 5597(1965).

36. J. P. Bennett, private communication.

37. D. A. Backingha-n, C. E. Davis and A. I1. Sargeson, J. Amer.
Che.. Soc., .2, 6159(1970).

38. F. Basolo and R. G. Pearson, "Mechanisms of Inorganic Reactions,"
2nd Ed., John Wiley and Sons, New York, N. Y., p. 32.

39. V. M. Kolthari and D. H. Busch, ILorg, Chem., 8, 5597(1965).

40. R. C. Weast, Ed., "Handbook of Chemistry and Physics," The
Chemical Rubber Publishing Company, Cleveland, Ohio, 1969,
p. D-109.

41. R. E. Hamm, R. L. Johnson, R. H. Perkins and R. E. Davis, J.
Amer. Chem. Soc., 80, 4469(1958).

42. Hi. Taubo, J. Chem. Ed., 45, 452(1968).

43. J. A. Laswick and R. A. Plane, J. Amer. Chem. Soc., 81, 3564
(1959).
W44. (a) C. W. Meredith and R. E. Connick, Abstracts of 149th Amer.
Chet. Soc. Meeting, Detroit, Michigan, Paper 1061, 1969.
(b) M. Eigen, Ter. Bunsenges. Phys. Chen., 67, 753(1963).
(c) R. C. Patel, R. E. Ball, J. F. Endicott and G. Hughes,
Tnorg, Choe., 2, 23(1970).
'5. R. H. Lane and L. E. Bennett, Chem. Coa., 191, p. 576.




103


46. Michael J. Gilroy, Ph. D. Thesis, University of Florida, 1971.

47. L. E. Bennett, R. H. Lane, M. J. Gilroy, F. A. Sedor,
P. F. Eisenhardt, E. L. Coomrbs and J. P. Bennett, to be
submitted.

48. D. H. Huchital and H. Taube, Inorg Chre., 4, 1660(1965).

49. (a) T. W. Swaddle and E. L. King, ibid., 532(1965).

50. J. Hine, "Physical Organic ChurAistry, 2nd Ed., McGraw-Hill,
New York, N. Y., 1962, pp. 169.













BIOGRAPHICAL SKETCH


Frank Alexander Sedor was born in East Chicago, Indiana,

on November 21, 1944. He graduated from Washington High School

of East Chicago in 1962, and received the Bachelor of Arts degree

with a major in chemistry from Wabash College, Crawiordsville,

Indiana.

He began his graduate studies in chemistry at the University

of Florida in September of 1966 where he held a teaching assistant-

ship until July of 1967. To September of 1968 he was supported by

a National Science Foundation grant, followed by a teaching a.ssis-

tantship until June of 1969, then an Interim Instructorship until

June of 1970, followed by a research assistantship to March of 1971,

then a teaching assistantship until August of 1971.

Mr. Sedor is married to the former Judith Anne Dye of Craw-

fordsville, Indiana, and is the father of a daughter, Julia

Christine, born, April 2, 1969.

In June of 1970 he received a DuPont Award for Excellence in

Teaching. Mr. Sedor is a member of the Am.erican Chemical Society.








I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




R. Carl Stoufer, Chairman
Associate Professor 6f Chemistry



I certify that I have read this study and that in ry opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Richard 1. Dresdner'
Professor of Chesistry



I certify that I have read this study ajnd that in Wm opinion it
confonrs to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissort-.tion for the degree of
Doctor of Philosophy.


A.. *. 'K o 9
Gus J. Palenik
Associate Professor of Chemistry
for
George E. Ryschkcritsch
Professor of Chce story



I certify that I have read this study and that in ry opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Jates A. Diiyrup
AsSociajte Profc-.cor of Chemistry









I certify that I have read this study and that in my opinion it
confonus to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Henry C. Brown
Professor of Chemical Engineering



This dissertation was submitted to the Department of Chemistry in
the College of Arts and Sciences and to the Graduate Council, and was
accepted as partial fulfillment of the requirements for the degree of
Doctor of Philosophy.

December, 1971




Dean, Graduate School















The IRctsio~sio of Thi.olato, A1koride, Thloet-hc.i and
Carboxylate tChelate Ur c .exs of Cobrit.( Ifl)
with Cromium(l1)












By

FRANK ALEKXL4 T' .R SE OR


A DIESERiTATION Ph'iEE:rEMD TO i',t GRPIUATE COUNCIL OF
THE UNIV,,'.:i' f ,F FL CiJ2. i' PARTIJl
FULFILLMENT OP T'i ... * I :S'' FOR THE 1'AEEE OiF









i': VEI3T O' 'l:iTO DA
1971
























,4Wt''


UNIVERSITY OF FLORIDA

3II 1262 0 ll552 4766
3 1262 08552 4766
















D1DICATION



To the i's














ACKNOWLEDGEMENTS


The author is deeply indebted to his research director,

Dr. L. E. Bennett, for his excellent guidance throughout the course

of this study. AppreciaLion is also expressed to the members of

t-ie -- 7 Cormitteo, Drs. R. C. Stouter, R. D. Dresdner,

OG J, Palenj!-, J, A. De yNn, and H. C. ir--,rn for their tire.

; .irac -. I velu counl.

Appreciation must bs expressed to the author's peers for

c .en:i,'S discussion bou`t related to and -ep at. front. academe.

By nroMe: IKessrs. P, F. Eicenhardt, M. J. Gilroy, S, R. Well.e-

and JJr. R. H. iane.

Deserving of most acknowledgement, however, are those members

of the author's immediate family for their forbearance of the author.














TABLE OF COIT:rTS


ACKNOIWEDG I ETS . . . .

LIST OF TAPES . . . . .

LTST OF FIGURES. . . . .

A STRACT . . . . . .

IMITRODUCTION . . . . .

EXPhPIuEIoTAL ...........

Materials. . . . .

Preparation of Comolexes

Ana ses . . . . .

Apparatus. . . . .

Treatment of Kinetic Data.


P.ESULTS . .


Characterization of Comple:

Reactions of Chnrium(II)


Ss. .

,n.th the


Complexes. . . . . . . . .

Activation Paranciccr for the Reaction Between the
Cobalt(III) Complexes and ChroaLium(II) . . .

Substitution Reactions of the Chrondium(III) Products

. . . . . . . . .. ....

Sc '1o.,n of th' C-(III) Cc.:;:es -b CIr:-nC(.LI) .

Che f +. CarboY ''.e as a ridin.r.r Fu'-or c . .

Chelt 3 10io: 2 as a Pi' 'i:g "un ......


Page
iii

vi

vii

viii

1

10

10

12

16

18

22

26


Cnbrit('CLI)


. .







TABLE OF COUIEYTS, continued



Page
Thiolate and Alkoxide as Bridging Functions. ... 87

Substitution Reactions at Chromiuri(III). ... .. .. 91

S ay. .............. . . . . 9

LIST OFr PEF0RECES .................. .. .100

BIOGAICAL SKETCH. . . . . . . . ...... 104













LIST OF TABLES


TABLE Page

I. Elution CharacterisLics of Some Cobalt(III) and
Chromi un(III) Species . . . . . . . . 27

II. Spectral Parameters of Some Cobalt(III) Species . 28

III. Spectral Parameters of Some Chromium(III) Species 29

IV. PMR of Various Protons Adjacent to Chalcogens . . 30

V. Stoichiom-tries of the Reactions of Cobalt(III)
Complexes with Cbromium(II) .. . . . . .. 49

VI. Acid r-ii Chroniw (iLI) Dependencies of the Reactions
of Some Cobalt(III) Complexes w.th Chromium(II) . 51

VIT. Acid : r C'-roi (I) Ti rec;-,rc of the Rizcrtion of
LCo(en)2(O0CCH (C3 )OH)J-' by GCromium(II) .....53
VIII. Rate Constants for Reactions of Co(III) Complexes
with Chromium(II) . . . . . . . . . 58

IX. Activation Parameters for the Reactions of
Chromium(II) with Cobalt(III) Ccmplexes ...... 64

X. Evaluation of Equilibrium Constant for the
Chroni-",(TII)-2-.'crr.aptopropionate Interconversion. 69

XI. Acid Dlci;ndence for Interconversion of the
Chrominux(III)-2 -::ercaptopropionate . . . ... .70

XII. Reactivity Parameters for the Reaction of Some
Cobalt(III) Complexes with Chrom5um(II) ....... .. 81











LIST OF FIGURES


FIGURE Page
1. Acid dependence of the reduction of [Co(en)2(OOCCH-
(CH )0H)]2+ by chromim(II). . . . . . .... 54
2. Eyring plot for [Co(en)g(OOCCH( 3)S)]. . . . 61
3. Eyring plot for [Co(en)2(0OCCH2SCH 3)]... . . 62
4. Eyring plots for some cobalt(III) complexes . . 63
5. Acid dependence of ring closure of [Cr(H!20)5(SCII(C3 )
COOH)] ...........2+. .. . . ... 67
6. Acid dependence of ring opening of [Cr(H20)4(0OCCH-
(CH3)S) . . . . . . . .. . ... 71
7. Acid independence of ring closure of [Cr(H20) (OOCCH-
(CH )SH)]2+ . . . . . . . 72







Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

TrE REACTIONS OF THIOLATE, ALKOXIDE, THIOECIP. AND
CAIROXYLATE CHELATE COM4PLEXS OF COBALT(III) I'TH CHROMIUM(II)

By
Frank Alexander Sedor

December, 1971

Chairman: R. C. Stoufer
Major Department: Chemistry

The reductions of 2-mercaptopropionatobic (ethylenedianine) cobalt-

(III) (A), lactatobis(ethylenedianine)cobalt(III) (B), 2-methyl-

thioacetatobis(ethylenedir ine )cobalt(III) (C) and glycinatobis-

(ethylencdiamine)cobalt(III) (D) by chromiun(II) were investigated

in an effort to asctrtain Jte contributions of sterically hindered

mercvptide and alke1 'e, terminal thiceth+er and chlate d ctrbrOy-

late functions, respectively, to inner-sphere reactivity .ara-

meters. For all reactions the stoichiometry was found to be

equimolar in oxidant and redactant. Product analysis indicated

all reactions proceeded via an inner-sphere pathway.

The reduction of (A) by chromiui(II) yielded a second-order

rate constant of 1.55 x 10 lo 1 see1 250C, an activation enthalpy

of 1,1 kcal/molo and an activation entropy of -31.1 eu. The ini-

tial product of the redox reaction was forlmlated as 2-mercaptopro-

pionatopentaaquochromium(III) (mercaptide-bound) (E). (E) was

observed as a fleeting intermediate and converted at an observable

rate to 2-mercaptopropionatotetraaquochro;.u (III) (chelated) (F)

(k = (1.0 + 2.6[H+]) x 10-2 (M and sec)), Chelate ring cloue

is two-fold faster than the analogous system with nrrcaptoacetete.







The chelato (F) equilibrated with a species characterized

as 2-mercaptcpropionatopentaaquochromium(III) (carboxylate-

bound) (G) (Ke = 10.5). Rate constants for ring opening and

closure were found to be kf = (7.31[H+]) x 10- and kr = 7.10

x 10-6 (M and sec). Comparison to the substituted chelate reveals

no significant difference in rates.

Reactivity parameters for the reaction of (B) with chromiium-

(II) were found to be k = (7.31 + 0.023[J+]-1 / 1 + Ka[H]-) and

k2 = 52 (1 and sec), AH* = 4.7 kcal/nole, AS* = -39.1 eu ([H+] =

1.0 Y). Methyl substitution on the chelate ring was found to

decrease the second-order rate constants equally for both mercap-

tide and alkoxide bridging functions compared to the unsubstituted

complexes previously studied.
Reaction of (C) Iwith chrcmiu m(II) yielded kobs = 267 (L and

sec, 250C),AHt = 8.5 kcal/mole and S"t = -18.9 eu. The product

of the initial redox reaction undern.ent at least one and perhaps

two observable subsequent reactions, indicating thioether as the

bridging function.

Reaction of (D) with chromiun(II) yielded k = 2.22 (L' and
-1
sec 250C),Ali* = 8.8 kcal/mole and AS = -27.4 eu. In com-

parison to monodentate carboxylate cobalt(III) complexes, the

presence of carboxylate in a chelate ring was concluded to enhance

the ability of the carbonyl function to serve as a bridging ligand.

Reactivity parameters obtained for (A), (B), (C) and (D) are

discussed relative to their mechanistic implications.




The Reactions of thiolate, alkoxide, thioethe and carboxylate chelate complexes of cobalt (III) with chromium (II)
CITATION SEARCH THUMBNAILS PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097688/00001
 Material Information
Title: The Reactions of thiolate, alkoxide, thioethe and carboxylate chelate complexes of cobalt (III) with chromium (II)
Physical Description: Book
Language: English
Creator: Sedor, Frank Alexander, 1944-
Copyright Date: 1971
 Subjects
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 001029710
oclc - 18111044
notis - AFB1812
System ID: UF00097688:00001

Downloads

This item has the following downloads:

( PDF )


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Tables
        Page vi
    List of Figures
        Page vii
    Abstract
        Page viii
        Page ix
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Experimental
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
    Results
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Discussion
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
    References
        Page 100
        Page 101
        Page 102
        Page 103
    Biographical sketch
        Page 104
    Back Matter
        Page 105
        Page 106
    Back Matter
        Page 107
Full Text










The Reactions of Thiolate, Alkoxide, Thioether and
Carboxylate Chelatte Complexes of Cobalt(III)
width Chromium(l1)











By

FRANK ALEXA'I)ER S3ORB


A LISSERTATIO; PI'.JE' TO "T!; GOrJ'UIAT" COUCIL OF
THE UlNViQSi'1'f ,F FLiTD'dI. i P'RTlAL
FULFILLMENT OF Trf: REQUITRiEV"' ;O:T: iO E D Ti REE OC
I C ()!u, F L.jlk(.Oh


UNI.VEfRSIT OF FIORTDA
1971














DEDICATION



To the J's













ACKNOWLEDGEMENTS


The author is deeply indebted to his research director,

Dr. L. E. Bennett, for his excellent guidance throughout the course

of this study. Appreciation is also expressed to the members of

the Supervisory Committee, Drs. R. C. Stoufer, R. D. Dresdner,

G. J. Palenik, J. A. Deyrup, and H. C. Brown, for their time,

patience and valued counsel.

Appreciation must be expressed to the author's peers for

enlightening discussions both related to and separate from academe.

By name: Messrs. P. F. Eisenhardt, M. J. Gilroy, S. R. Weller

and Dr. R. H. Lane.

Deserving of most acknowledgement, however, are those members

of the author's immediate family for their forbearance of the author.














TABLE OF CONTENTS


ACKNOWL ENECTS . . . . . . .. . . . .

LIST OF TABLES .......................

LIST OF FIGURES .....................

ABSTRACT..........................

INTRODUCTION .. . ........... .. .

EXPERIMENTAL .. . .......... . ....

Materials. .....................

Preparation of Complexes . . . . . . . .

Analyses .......................

Apparatus. . . . . . . . . . ... .

Treatment of Kinetic Data. . . . . . .

RESULTS.. . . . .............. .

Characterization of Complexes. . .. . .

Reactions of Chromium(II) with the Cobalt(III)
Complexes................... ...

Activation Parameters for the Reaction Between the
Cobalt(III) Complexes and Chromium(II) .. . . .

Substitution Reactions of the Chromium(III) Products .

DISCUSSION ....................... ..

Reduction of the Co(III) Complexes by Chromium(II) ..

Chelated Carboxylate as a Bridging Function. . . .

Chelated Thioether as a Bridging Function. . . .


Page
iii

vi

vii

viii

1

10

10

12

16

18

22

26

26


47








TABLE OF CONTENTS, continued


Page
Thiolate and Alkoxide as Bridging Functions . . 87

Substitution Reactions at Chrorium(III). . . .. 91

Su . . . . . . . 98

LIST OF REFERNCES ................ ... 100

BIOORAPHICAL SKET~I. . . . . . . . 10*













LIST OF TABLES


TABLE Page

I. Elution Characteristics of Some Cobalt(III) and
Chromium(III) Species . . . . . . . . 27

II. Spectral Parameters of Some Cobalt(III) Species .. 28

III. Spectral Parameters of Some Chromium(III) Species . 29

IV. PMR of Various Protons Adjacent to Chalcogens . . 30

V. Stoichiometries of the Reactions of Cobalt(III)
Complexes with Chromium(II) . . . . ... 49

VI, Acid and Chromium(II) Dependencies of the Reactions
of Some Cobalt(III) Complexes with Chromium(II) . 51

VII. Acid and Chromium(II) Dependence of the Reduction of
[Co(en)2(OOCCH(CH3)OH)?" by Chromium(II) . .. 53
VIII. Rate Constants for Reactions of Co(III) Complexes
with Chromium(II) ..... ... ......... . 558

IX. Activation Parameters for the Reactions of
Chromium(II) with Cobalt(III) Complexes . . . 64

X. Evaluation of Equilibrium Constant for the
Chromium(III)-2-Mercaptopropionate Interconversion, 69

XI. Acid Dependence for Interconversion of the
Chromium(III )-2-4:ercaptopropionate. . . . ... 70

XII. Reactivity Parameters for the Reaction of Some
Cobalt(III) Complexes with Chromium(II) . . .. .81












LIST OF FIGURES


FIGURE Page
1. Acid dependence of the reduction of [Co(en)2(OOCCH-
(CH 3)OH)f+ by chromiun(II). . . . . ... ..
2. Eyring plot for [Co(en)2(OOCCH(3)S) . . . .. 61
3. Eyring plot for [Co(en)2(OOCCH2SCH 3)j. . . ... 62
4. Eyring plots for some cobalt(III) complexes. . . 63
5. Acid dependence of ring closure of [Cr(H20)5(SCH(CH3)
COOH)]2+ ....................... 67
6. Acid dependence of ring opening of [Cr(H20)4(OOCCH-
(C3)S)J . . . . . . ... . . .. 71
7. Acid independence of ring closure of [Cr(HO)5(OOCCH-
(CH)SH)])f+ ................. .....








Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE REACTIONS OF THIOLATE, AIlOXIDE, THIOEIER AND
CARBOXYLATE CHELATE COMPEXES OF COBALT(III) WITH CHROMIUM(II)

By

Frank Alexander Sedor

December, 1971

Chairman: R. C. Stoufer
Major Department: Chemistry

The reductions of 2-mercaptopropionatobis(ethylenediamine)cobalt-

(III) (A), lactatobis(ethylenediamine)cobalt(III) (B), 2-methyl-

thioacetatobis(ethylenediamine)cobalt(III) (C) and glycinatobis-

(ethylenediamine)cobalt(III) (D) by chromium(II) were investigated

in an effort to ascertain the contributions of sterically hindered

mercaptide and alkox-de, terminal thicether and chelated carboxy-
late functions, respectively, to inner-sphere reactivity para-

meters. For all reactions the stoichiometry was found to be

equimolar in oxidant and reductant. Product analysis indicated

all reactions proceeded via an inner-sphere pathway.

The reduction of (A) by chromium(II) yielded a second-order

rate constant of 1.55 x 105 M- sec", 250C, an activation enthalpy

of 1.1 kcal/mole and an activation entropy of -31.1 eu. The ini-

tial product of the redox reaction was formulated as 2-mercaptopro-

pionatopentaaquochromium(III) (mercaptide-bound) (E). (E) was

observed as a fleeting intermediate and converted at an observable

rate to 2-mercaptopropionatotetraaquochromium(III) (chelated) (F)

(k = (1.0 + 2.6[1I']) x 102 (N and sec)). Chelate ring closure

is two-fold faster than the analogous system with mercaptoacetate.







The chelate (F) equilibrated with a species characterized
as 2-mercaptopropionatopentaaquochromium(II) (carbozylate-
bound) (0) (K = 10.5). Rate constants for ring opening and
closure were found to be kf (7.31H]) x 10-5 and r = 7.10
x 10 (M_ and sec). Comparison to the substituted chelate reveals
no significant difference in rates.
Reactivity parameters for the reaction of (B) with chromium-
(I) were found to be k (7.31 + 0.023IHI" / 1 + Ka[H+"1) and
2 = 52 (M and sec), AH* = 4.7 koal/mole, AS* = -39.1 eu ([H] =
1.0 1M). Methyl substitution on the chelate ring was found to
decrease the second-order rate constants equally for both mercap-
tide and alkoxide bridging functions compared to the unsubstituted
complexes previously studied.
Reaction of (C) with chrosmum(Il) yielded ko = 267 (g and
seec, 25*C),AH* = 8.5 kcal/mole and AS = -18.9 eu. The product
of the initial redox reaction underwent at least one and perhaps
two observable subsequent reactions, indicating thioether as the
bridging function.
Reaction of (D) with chromium(II) yielded k = 2.22 (Q and
see"1, 25oC),AH* = 8.8 kcal/mole andAS* = -27.4 eu. In com-
parison to monodentate carboxylate cobalt(III) complexes, the
presence of carboxylate in a chelate ring was concluded to enhance
the ability of the carbonyl function to serve as a bridging ligand.
Reactivity parameters obtained for (A), (B), (C) and (D) are
discussed relative to their mechanistic implications.














INTRODUCTION


Although the coordination of metal ions by sulfur in the

form of mercaptide, sulfide, or thioether functions have long

been known, only during the last decade have their interactions

been recognized as vital in biological systems. Metal-sulfur

coordination is now known to play key roles in the processes of

photosynthesis, nitrogen fixation, oxygen metabolism, hydroxy-

lation of steroidal compounds and electron transport. Also

several compounds whose specific biological function has not yet

been elucidated are known to have metal-sulfur coordination.

The most explicit demonstration of metal coordination by

sulfur has come from recent structural determinations by x-ray

crystallography. In clostridial rubredoxin, whose function is

unknown, the single iron atom is coordinated tetrahedrally by four

cysteinyl mercaptide functions. Spectral studies indicate that

this coordination is retained in solution.3b In horse heart and

bonito cytochrome c, the heme iron is coordinated in the out-of-

plane positions by imidazole nitrogen and methiogrl thioether
4
functions. Thus the importance of both mercaptide and thioether

coordination is definitely established.

Extensive chemical and physical investigations to date have

been carried out on members of the class of non-heme iron proteins

(NBIP) known as ferredoxins which are involved in photosynthetic








and nitrogen-fixation processes.1' The isolation of several

proteins of this type has formed the basis of investigation work.

In general, the compounds of this class contain stoichiometrically

related non-heme iron, cysteine, and acid-labile sulfur, exhibit

physical parameters which are anomalous for iron and serve in

biological electron transport functions. Specific familial

characteristics of ferredoxins are (1) their relatively low

molecular weight (-12,000-30,000 g/mole), (2) the presence of acid-

labile sulfur (treatment with acid produces HZS) in an amount

approximately equivalent to their iron content, (3) a stoichio-

metric relationship of iron and cysteinyl ligands, (4) electronic

absorption and electron paramagnetic resonance which is anomalous

for iron complexes and (5) oxidation potentials (0.2 to 0.4 v at

pH = 7) which are unprecedented for iron complexes. Chemical

subonits of this class are also incorporated, with other redox

functions, into more complex enzymes which are utilized in various

biological redox processes.1

The presence of both iron and labile sulfur at the active

site of a typical two-iron protein was established by epr measure-

ments on isotopically substituted species.2b Iron and acid-labile

sulfur were removed from the protein to yield the inactive apo-

protein. Biological activity was restored by treatment with iron

salts in combination with 2-mercaptoethanol and inorganic sulfide.

Substitution of selenium for inorganic sulfur also regenerated a
7
substantially active protein. The splitting of the epr signal

in the reduced protein by independently substituted appropriate

isotopes of iron and selenium established their mutual proximity








to the site of reduction. A similar proximity for the oysteinyl

sulfur atoms is suggested by the epr behavior of protein produced

by organisms grown on an isotopically substituted source of

sulfur.7

This brief summary is representative of recent research

which demonstrates an extensive and varied utilization of metal

ions coordinated by thiolate, thioether and "labile" sulfur atoms

in biological oxidation-reduction processes. The research to be

described here represents an effort to examine the influence of

two of these donor functions on the redox behavior of certain

metal complexes. The complexes were chosen not for the extent to

which they simulated the biological examples but rather for their

virtue of incorporating certain of the biological aspects into

systems whose reactions stood the best chance of being both

thoroughly characterized and interpreted in relationship to prior

fundamental studies. This objective stands in contrast to the

alternative approach of investigating systems more directly related

to those found biologically, However, the iron(III)mercaptide

complexes suggested by the biological systems are complicated,

unstable and kinetically labile, leading to less than definitive

results. Simple complexes with iron-thioether coordination have

proved elusive to synthesis and would likely present a low suscep-

tibility to thorough kinetic description.

The reactions described herein serve as model systems in

only the most rudimentary sense that they do incorporate the bio-

logical type of donor functions as ligands. The possibility of

substantial differences between the reactivity patterns delineated








here and those of the biological examples is openly anticipated;

nevertheless, a better understanding of how these ligands affect

oxidation-reduction behavior should shed some light on the reasons

for their extensive biological utilization.

One approach to evaluating the effects of mercaptide and

thioether coordination involves the study of well-characterized

complexes whose reactions are susceptible to detailed mechanistic

investigation. Further, monomeric, inert complexes of cobalt(III)

or chromium(III) containing a single coordinated mercaptide or

thioether function appeared desirable in the initial studies for

coapaative purposes with earlier work. With these objectives, a

research project in these laboratories resulted in the preparation

and isolation of [Co(en)2(00CCH2S)]cD40 9 For the purpose of

comparing the reactivities and mechanistic patterns an analogous

compound with the sulfur replaced by a more classical oxygen donor

atom was prepared, [Co(en)2(OOCCH20)]C104. For the pair of com-

plexes, the behavior toward chromium(Il) and [Ru(NH 3)6P as

typical inner-and outer-sphere reductants, respectively, as well

as applicable substitutional behavior of the reaction products

was examined.9'10 This marked the first significant development

in evaluating the influence of a coordinated mercaptide in oxida-

tion-reduction reactions.

In order to consider these results and those described

herein it is necessary to recognize the two general categories

which have been established for oxidation-reduction reactions of

transition metal complcxcs in solution, that is, the inner-sphere

and outer-sphere mechani;,s.l For the inner-sphere reaction, the








two reacting metal centers are joined in the activated complex

via a ligand common to the first coordination spheres of both

metal ions. An outer-sphere reaction is characterized by an

activated complex in which no sharing of ligands occur, that is,

the primary coordination spheres of the respective metal ions

remain intact with no bond cleavage or bond making required for

electron transfer.

For an inner-sphere reaction one of the reactants must be

sufficiently labile so that ligand substitution involving a ligand

from the second reactant can occur prior to electron transfer. The

bridging ligand(s) of the second complex is required to have

available electron pairs or orbitals of sufficient energy and

proper symmetry for forming a bridge between the two metal centers.

The bridging ligand brought in by the second reactant must be re-

tained long enough for the electron transfer to occur. By meeting

the above requirements an inner-sphere reaction is made possible

but not mandatory. An outer-sphere reaction may still provide a

path of lower energy.

A reaction is most conveniently assigned to the inner-sphere

category if the bridging ligand can be detected in the coordination

sphere of the product of the labile reactant at higher than equil-

ibrium levels. This requires that it be retained due to an inert-

ness of this product to substitution relative to the rate of the

redox reaction. Exemplifying this means of categorization is the

classical reaction between the substitution-inert complex, [Co(N-

H)5(C1)]2+, and substitution-labile [CG0)62+ in acid solution

to yield NH substitution-labile [Co(H20)61 and substitution-







inert [Cr(H2o)5(cl)f2 .12 The capture of the chloride ion in

the first coordination sphere of the inert chromium(III) product
is definitive of an inner-sphere reaction. When the same

reaction is carried out in the presence of radioactive free

chloride ion, no radioactivity is found in the chromium(III)

product, thus eliminating its incorporation prior or subsequent to

electron transfer which identifies the coordinated chloride in the
12
product as originating with the cobalt(III) complex. Other

donor functions which have been identified as bridging ligands for

cobalt(III)-chromium(II) reactions are the halides, carboxylates,

asides, thiocyanates, phosphate, sulfate and hydroxide,1213 In

contrast, an outer-sphere reaction is decisively dictated if one

of the reactants is substitution-inert relative to the rate of

electron transfer and contains in its first coordination sphere

no ligand capable of bridge formation. A well-characterized

example is the outer-sphere reductant, [Eu(I3 )6 2.
The investigation of the mercaptide and alkoxide complexes

previously described established, through characterization of the

chromium(III) products, that the reactions of the species with

chromium(II) occur via inner-sphere paths.10 The reduction of

[Co(en)2(o0CCH2o0) was concluded to occur via a bridging reaction

utilizing the alkoxide oxygen from kinetic comparisons. It is

directly comparable, therefore, to the reduction of [Co(en)2(OOC-

CH2S)] which proceeded via a mercaptide bridged path as esta-

blished by product characterization. The results clearly esta-

blished a reactivity toward chromium(II) for the sulfur complex

which is -103 times greater than for its oxygen analogue.








While the previous research established that the coordinated

mercaptide exerted a substantial increase in reactivity for oxida-

tion-reduction, several important questions remained unanswered.

The various factors which might possibly contribute to thisenhsnoed

reactivity remained unevaluated though recognized. In particu-

lar, the relative influence of steric vs electronic factors could

not be determined. It seemed desirable to evaluate these influ-

ences in complexes as similar as possible in order to minimize

any effects arising from the contribution of the standard free

energy change to the reactivities.15 Thus, if a greater steric

accessibility of the larger sulfur atom were mainly responsible for

the observed reactivity pattern in otherwise comparable complexes,

it was felt substitution of a methyl group for one of the adjacent

methylene protons should decrease the rate reduction of the sulfur

complex less than that of the analogous oxygen complex. Further,

if the anticipated rate decrease is observed, the possibility

would arise of determining the enthalpies and entropies of activa-

tion which were not accessible for the mercaptoacetato complex.

These should shed further light on the steric and electronic con-

tributions to the reactivity differences. Thus, the complexes

[Co(en)2(OOCCH(CH3)S)](C104) and [Co(en)2(OOCCH(CH3)O)](C104) were

prepared in order to further define the influence of coordinated

mercaptide.

It was hoped that the above modification in the alkoxide

ligand would permit a more definitive characterization of the

chromium(III) product as deriving from alkoxide bridging than was

possible with the previously investigated analogue. When this was








found not to be the case, a complex having a chelate ligand which

contained a carboxylate function as the only possible bridging

group, [Co(en)2(OOCC2i2,2)] was investigated in order to iso-
late any unique effects arising from chelation relative to those

previously established for monodentate carboxylate coordination.1

Comparison utilizing results with this complex were expected to

more rigorously define, and possibly exclude, the participation

of carboxylate bridging for the alkoxide-containing complexes.

In spite of its demonstrated biological importance the redox

influence of coordinated thioether functions remains poorly under-

stood. The first kinetic study directly involving this mode of

coordination in simple systems suggests that the effect of thio-

ether donor functions as non-bridging ligands in an inner-sphere

reaction is to enhance reactivity.17 Earlier rate studies of coa

plexes such as [Co(NH3)5(0OCC2sca 2C6H )] suggested that a pendant

thioether enhanced electron transfer rates above that observed

with the coordinated carboxylate alone, presumably ia chelati

of the reductant. In anticipation of this possibility and in

collaboration with concurrent research1 the complex [Co(en)2-

(00CCH2SCH3)1+ was prepared and studied. Whatever the mechan-
istic pathway taken, it was expected that the similarity of this

complex to the previously studied mercaptide precursor should

provide a better understanding of the influence of coordination

by a thioether donor.

Thus, the objectives of this thesis include (1) evaluation

of the steric and electronic factors responsible for the increased

reactivity of mercaptide complexes relative to their oxygen ana.




9



logues, (2) further comparison of these influences within the

respective alkoxide and mercaptide classes of complexes, (3)

definition of the role played by carboxylate incorporated in

chelate system as opposed to the monodentate carboxylate function

and (4) an investigation into the accessible aspects of thioether

coordination with the additional complex made available by this

research. Further, the behavior of the chromium(III) complexes

uniquely produced by reactions originally investigated for their

relevance to oxidation-reduction chemistry should contribute to

a better understanding of the substitutional characteristics of

these ligands.














EXPERIMENTAL


Materials



Reagents.--Common chemicals were of reagent grade and were

used without further purification unless otherwise specified.

Water.--Distilled water used in kinetic experiments and in

preparation of various stock reagent solutions was obtained by the

distillation of deionized water from alkaline permanganate solu-

tion using an all-glass distilling assembly and stored in poly-

stoppered, glass bottles.

Nitrogen.--For deaerating solutions of air-sensitive materi-

als, line nitrogen was passed through two successive scrubbing

towers containing Cr(II) solutions which were prepared by reduc-

ing 0.1 M chromium(III) ion in 0.4 M perchlorio acid in the towers

with a bed of amalgamated zinc. To assure the nitrogen was satur-

ated with water, it was then passed through a tower containing

redistilled water.

Zinc amalgam.--Twenty mesh granular zinc (Fisher) was activa-

ted with 3 M HCL. After several rinsings with distilled water the

activated zinc was amalgamated using a solution of tetrachloro-

mercurate(II) ion in 1 M HC1 for ten seconds. After several wash-

ings with distilled water the amalgam was dried under a stream of

dry nitrogen.








Lithium perchlorate.--Reagent grade lithium perchlorate was

used throughout to maintain constant ionic strength. Purity was

checked by passing a prepared solution through an ion exchange

column in the acid form and titrating the collected solution with

0.1 M sodium hydroxide to a phenophthalein end point.

Chromium(II) ion.--Reagent grade chromium(III) perchlorate

(G. F. Smith) was used to prepare stock solutions ranging in con-

centration from 0.050 M to 0.25 M. Aliquots of the stock solutions

were diluted to concentrations ranging from approximately 4 x 10-

M to 2 x 10-1 M in solutions varying in acidity from 1 x 104

to 1 1j. Reduction to chromium(II) was accomplished using zinc

amalgam.

Chloropentaamminecobalt(II) chloride.--This compound was

kindly made by Mr. Peter F. Eisenhardt, who used a standard pro-

cedure for its preparation.20 The compound exhibited a molar

absorbtivity of 49.9 at 534 nm in excellent agreement with the

values of 50.2 51.0 previously reported.2122

Mercaptoacetatobis(ethylenediamine)cobalt(III) verchlorate.-

This compound was generously provided by Dr. Robert H. Lane, by

whom it was first prepared and characterized.9 The sample provided

exhibited the following spectral characterization: [A(E): 518(1521

282(11,700)].

2.2'-Dithiodioionic acid.--2-mercaptopropionic acid

(Aldrich Chemical Co., reagent grade) was converted to the disul-

fide by a modification of the method of Fredga and Bjorn23 which

appears to be general for preparation of simple disulfides from the

corresponding mercaptans.8def 24 One millimole of the mercaptan








was slowly added to a solution of one millimole iron(III) per-

chlorate dissolved in 200 ml of water. After addition the solution

was allowed to stir for one hour, several drops of concentrated

sulfuric acid was added, and the reaction mixture extracted with

three 50 ml portions of diethyl ether. The combined ethereal

extracts were evaporated to dryness under a stream of nitrogen.

The resultant white solid exhibited a melting point of 1130C

uncorrected (lit. 1140 5.0C)23 and was used without further

purification.


Preparation of Complexes


2-Mercartopropionatobis(ethylenediamine)cobalt(III) Perchlorate--
Co(en)2(OOCCH(CH )S) JC104

Preparation of this compound paralleled the method used in

preparation of the mercaptoacetatobis(ethylenediamine)cobalt(III)

complex previously reported.9 Ten grams (.028 mole) of Co(C104)2.
6H20 (G. F. Smith) was dissolved in 30 ml of H20 in a 100 ml

three-neck flask fitted with rubber septum stopper, nitrogen inlet

and outlet, and magnetic stirrer. After deaeration for thirty

minutes, 3.6 ml (.057 mole) of 98-100 per cent ethylenediamine

(Baker) was added by syringe. After another thirty minutes deaer-

ation time, 2.85 g of solid 2,2'-dithiodipropionic acid was added
by rapid removal and replacement of the nitrogen outlet tube.

Within five minutes the solid had dissolved and the solution color

changed slightly to a brown color. After thirty minutes under

nitrogen the solution was transferred to a 150 ml beaker, and







evaporated with stirring on a hot plate at 1000C under a gentle

stream of nitrogen until the volume was approximately 20 ml then

allowed to cool to ambient temperature. The resultant foayr brown

mass was filtered and washed with 5 ml of hot water. Recrystalli-

zation was effected by dissolving the collected solid in 100 ml of

hot (90C) water, filtering, and cooling in an ice-acetone slush.

Light purple needles separated on cooling. The product was re-

crystallized twice more as described, washed with two 10 ml por-

tions of absolute ethanol and one 10 ml portion absolute diethyl

ether, then dried in vacuo over CaSO4 for twelve hours. Yield =

2.5 g. Anal. Calcd, for [Co(mN2CH2CH2NH2)2((CCH(CH3)S)]Cl04:

C, 21.97; H, 5.27; N, 14.64; S, 8.36; Co, 15W0. Found: C, 21.75;

H, 5.34; N, 14.61; S, 8.32; Co, 15.15.

Lactatobis(ethylenediamine)cobalt(III) Perchlorate-4Co(en),21(OCC-
(CH 3)0)JCl04

Eighteen grams (.05 mole) of Co(C104)2g6H20 (G. F. Smith)

was dissolved in 125 ml of water in 250 ml Erlenmeyer flask fitted

with a two-hole rubber stopper with glass tubing of appropriate

lengths. Six grams (.10 mole) ethylenediamine (Baker) was added

with stirring, then air was drawn through the solution for twelve

hours. To this solution was added a solution of 5.05 g (.05 mole)

70% aqueous lactic acid (Baker) and 2.00 g (.05 mole) NaOH pellets

(Fisher) in 20 ml water. The combined solutions were transferred

to a 250 ml beaker and evaporated under a gentle stream of nitro-

gen with stirring on a hot plate at 1000C until the volume was

approximately 100 ml (ca. 40 min). It was cooled to ambient

temperature and the solids present were filtered and washed with







10 ml water. The combined filtrate and washing was returned to
the hot plate and evaporated as before to a volume of 50 ml,
then cooled to ambient temperature. The resulting solid was
filtered, washed with 5 ml of water, 15 ml of absolute ethanol,
and 15 ml of absolute diethyl ether and dried in vacuo for ten
hours over CaSO4. Yield = 4.75 g. Anal. Called. for [Co(HN2-

CH2CH2NH2)2(OOCCH(CH3)0)IC104: C, 22.90; H, 5.52; N, 15.27;
Co, 16.11. Found: C, 22.86; H, 5.52; N, 15.14; Co, 15.82.
2-Methylthioacetatobis(ethylenediamine)cobalt(III) Diperchlorate--
[Co(en),(oocCHaSCHI )](clo4)2

To a suspension of 1.8 g of [Co(en)2(OOGC3)j in 300 ml
of 90% methanol-water mixture was added a large excess (25-fold)
of methyl iodide (Baker). The mixture was stirred in a closed,
round-bottom flask for twenty-four hours and for three hours was
stirred while open to the atmosphere. The light pink solid was
collected, washed with two 25 ml portions of absolute ethanol, two
25 ml portions of diethyl ether and dried in vacuo over CaSO for
twelve hours. The solid was recrystallized by dissolution in
minimum amount of hot (90C) water followed by an addition of
solid NaC104 (~4 g) until precipitation began. After allowing it
to cool to ambient temperature, the bright red-orange solid was
collected and recrystallized again. Yield = 0.70 g. Anal. Calod.
for [Co(NMH22CgC2N2)2(OOCCH2SCH3)](C104)2: C, 17.40; H, 4.38;
N, 11.60; S, 6.64. Found: C, 17.42; H, 4.26; N, 11.64; S, 6.75.








Glvcinatobis(ethylenediamine)cobalt(III) Diverchlorate-- Co-
(en)2(OOCCHI2NH2)J(C104)2

This compound was prepared by the reaction of silver perchlo-

rate with glycinatobis(ethylenediamine)cobalt(III)dichloride.1

Twenty grams of trans-[Co(en)2C12]Cl was suspended in 40 ml of

water. To this was added 8.1 g silver oxide (Matheson, Coleman,

and Bell) and the suspension was ground in a mortar and pestle

periodically for one hour. Silver chloride was removed by filtra-

tion and washed with 40 ml of hot (900C) water. To the combined

filtrate and washings was added 6.9 g glycine (Fisher), and then

the mixture was evaporated on a steam bath under a stream of nitro-

gen to a thick syrup. After standing at ambient temperature, the

resultant solid was filtered, washed with cold water, recrystallied

from hot water, and dried in vacuo over CaSO4 overnight.

A solution of 6.81 g (0.02 mole) of [Co(en)2(OOCCH2H2)Cl2

dissolved in 40 ml water was added slowly with stirring to a silver

perchlorate solution prepared by adding 2.36 g (0.0085 mole) silver
carbonate (Mallinkrodt) to 10 ml of 2 M perchlorate acid. Silver

chloride was removed by filtration and the filtrate was reduced on

a rotary evaporator at 450C to a thick syrup. After several days'
standing at ambient temperature the resultant solid mass was filteed

and air dried using aspirator suction for 2 days. The solid was

dried in vacuo over CaSO4 for twenty-four hours. Yield = 5.0 g.

Anal. Calcd. for [Co(N'2CH2CH2NH2)2(OOCCH2NHi2)](C104)2: C, 15.94;

H, 4.43; N, 15.49; Co, 13.04. Found: C, 16.13; H, 4.52; N, 15.52;

Co, 13.25.








Chromium(II) Complexes of 2-MercantorAonionic id L Acid
ad Glvcine

Several aquo-chromium(III) complexes with these ligands were

generated in solution by the reduction of the cobalt(III) complexes
with chromium(II). Some of the complexes so produced underwent

further substitutional changes, the products of which were separ-

ated according to charge type by ion exchange chromatography and

further characterized by their ultraviolet and visible spectrum

and by chromium analyses. Synthesis and isolation of the complexes

was not a primary objective; therefore, discussion of them is

deferred to the section on Results.



Analyses


Chromium(II ).--Determination of chromium(III) was accomplished

spectrally by alkaline peroxide oxidation to chromate(VI) ion which

was monitored at 373 rm ( = 4,815 15).25 To an aliquot of the

chromium(III) complex, 10 ml distilled water, 10 ml 0.20 I Na0H

and 3 ml of 30% hydrogen peroxide were added. The oxidation

usually was complete overnight. Excess peroxide was deoonposed

by heating the solution at 600C using a coiled platinum wire as

catalyst. When cool, the solution was diluted to 100 ml and

absorbance at 373 m was observed. Duplicate runs usually were

reproducible to within 1%.

Chromium(II).--Periodically aliquots of chromium(II) solu-
tions were reacted in an inert atmosphere with a solution contain-
ing a known excess of chloropentaamnnineoobalt(III) in 0.1 U HC104.

The excess was determined by analyzing spectrally the resultant








solution at 534 nm, after subtracting for absorbances due to

presence of hexaquocobalt(II) ion (E-,+ = 31)26 and chloropenta-

aquochroum(III) ion (E,-5 = 5.5).
Cobalt.-The method of Moss and Mellon using 2-2',2"-ter-
yridine as a completing agent was used.28 A sample of complex con-

taining approximately 4 mg cobalt was destroyed using 15 1m liquid

fire reagent (7 parts 70% HC104 3 parts 70% HN03). The reaction

mixture was evaporated just to dryness on a hot plate, with the
residue then dissolved in 20 ml distilled water. Addition of 5 ml

of 20% aauonium acetate solution brought the pH to about 6 and the

solution was then diluted to 100 ml. A 25.0 ml aliquot of the
cobalt(II) solution was deaerated using a 50.0 volumetric flask
fitted with septum stopper and hypodermic needles for nitrogen in-
let and outlet. Ten milliliters of 0.2% terpyridyl solution was

injected and volume brought to 50.0 ml by syringe addition of
* deaerated distilled water. A sample was then transferred to a
deaerated septum-stoppered 2 cm cell using syringe techniques and

the concentration of the cobalt(II) terpyridyl complex determined
spectrally at 505 m. A molar absorbtivity of 1,386 at 505 nm

determined with known cobalt solutions29 was used rather than the

reported value of 1,360.28
Elemental analyses.-Analyses for carbon, hydrogen, nitrogen

and sulfur were performed by Galbraith Laboratories, Inc.,
Knoxville, Tennessee. Nitrogen analyses were performed by the
Kjeldahl method.








Apparatus



It was necessary to exclude the presence of oxygen from

reactant solutions used in kinetic studies due to the extreme

sensitivity of the chromium(II) species to oxidation in solution.

This was accomplished by deaeration of all solutions using three-

neck flasks fitted with nitrogen inlet, outlet, and rubber septum

for syringe withdrawal of solutions. Times for purging ranged

from thirty minutes to one hour. Transfers of deaerated solutions

were nado using all-glass graduated syringes fitted with stainless

steel needles.

For all kinetic studies at least one of the reactants exhi-

bited a characteristic absorbance in the visible and/or near ultra-

violet region. Reactions were then monitored optically at the

respective absorbances.

Kinetic Studies

Fast reactions (1 msec < ti > 15 sec) were monitored using a

Durrum-Gibson Stopped-Flow Spectrophotometer equipped with tungsten

and deuterium light sources, Kel-F flow system, and an AMINCO

4-8600 external temperature bath. Use of an external circulating

pump maintained a constant temperature environment for the drive

syringes, mixing jet, and observation chamber.

Thermostatted, deaerated reaotant solutions were transferred

to the stopped-flow assemble via 20 ml reservoir syringes. The

drive syringes were filled from the reservoir syringes by a system

of externally sealed Kel-F valves. The filled drive syringes were

then allowed to equilibrate to the desired temperature (estimated)








for five to thirty minutes. A plunger armed by nitrogen pressure

(65 psi) and fitted with external, integrated trigger then forced

the reactant solutions into the observation chamber via the mixing

jet. Mixing and instrument mechanical dead times (2 msec) were,

in some instances, comparable to the shortest half-life of the

reactions so studied. For these cases it was necessary to allow

for the dead time and begin actual reaction study after one or two

half-lives had passed.

Data were recorded first on the storage oscilloscope, then

reproduced using Type 47, 3000 speed Polaroid Film (B & W) using

a Polaroid Land Camera bank mounted on a Tektronix C-27 Oscillo-

scope Camera with appropriate bezel attachment. Kinetic data were

obtained in the temperature range 15.00C to 45.0C as necessary.

No special precautions were taken to exclude oxygen other

than flushing the flow system with several volumes of the deaerated

reactant solutions. This technique was found to be sufficient for

solutions of chromium(II) ion concentration greater than 4 x 10

M as evidenced by reproducibility of the observed rate constant

within 10%) for the Cr(II)-[Co(en)2(OOCCH(CH3)S)1 reaction when

the concentration of Cr(II) prior to mixing was 4.3 x 10 M- .

Below this level partial consumption of the reductant occurred.

Slow reactions were followed by use of Gary 14 Recording

Spectrophotometer fitted with constant temperature cell housing.

Temperature was maintained with an AMINCO 4-8605 constant tempera-

ture bath equipped with an external circulating pump. Reaction

solutions were again deaerated and thermostatted before mixing.








Some of the reactions followed on the Cary 14 utilized a

glass mixing apparatus described pictorially elsewhere.14b

Essentially the apparatus consisted of two reservoirs placed so

as to form a "V" with an aperture for a rubber septum stopper to

introduce the respective reactant solution and with provision for

maintaining an inert atmosphere. At the apex of and perpendicular

to the "V" was delivery tube with pressure equalizing arm, fitted

with ground glass joint appropriate for the quartz cells.. Tilting

the apparatus, and shaking to mix, then draining into the cell

effected reaction. Initial data were obtainable within ten seconds

of mixing.

Spectronhotometry

Visible and near-ultraviolet spectra were obtained with a

Cary Model 14 Recording Spectrophotometer. Sample solutions and

baselines of pure solvent were run against air as the reference.

Quartz cells of 0.10, 1.00, 2.00, 5.00, and 10.0 cm path lengths

were available.

Infrared spectra were obtained from 4,000 to 625 cm- with

a Perkin-Elmer Model 337 recording infrared spectrophotometer.

All samples were run as KBr pellets.

Proton Magnetic Resonance Spectra

PMR spectra were obtained with a Varian A-60A Analytical IMR

Spectrophotometer with a magnet temperature of 37C. In order to

reach a concentration of complex large enough to obtain meaningful

spectra for the 2-mercaptopropionic acid complex, it was necessary

to convert the complex from the perchlorate to the chloride salt.

This was accomplished by the reaction of 0.100 mmole of complex








with 0.095 mmole of tetraphenylarsonium chloride (G. F. Smith) in

a minimum volume of water. The resulting tetraphenylarsonium

perchlorate formed was removed by filtration, and the filtrate

evaporated to dryness on a rotary evaporator and dried in a vacuum

desiccator overnight.

Spectra of saturated solutions of the chloride in deuterium

oxide (99.5%, Matheson, Coleman, and Bell) were obtained with TMS

used as an external standard. Solutions were made acidic (pD = 1)

with trifluoroacetic anhydride (Aldrich).

The solubility of the lactic acid and methylthioacetic acid

complexes posed no such problem and spectra were obtained using

the perchlorate salt.

Ion k..i.:iir,-e tlii ic

All ion exchange separations were carried out using Biorad

AG 50W-X2, 200-400 mesh (purified standard Dowex resin of the same

designation), analytical grade cation exchange resin. The resin

was converted to either the sodium or lithium form from the hydro-

gen form by soaking and washing the resin in a solution of 1 N in

NaOH--NaCIO4 or LiOH--LiC104, then rinsing with distilled water

until the eluent was neutral to Hydrion paper.

Because of the fine mesh of the resin, nitrogen pressure

(15 psi) was used to increase the flow rate from 1-2 ml/min to

6-7 ml/min. Separation of bands was adequate under conditions of

the latter flow rate. Elution characteristics for several ions of

interest are given in Table I (p. 27).







Treatment of Kinetic Data


For all the oxidation-reduction reactions studied it was
determined that the reactions were first order in both oxidant and

reductant over the ranges considered. The stoichiometry was
established to be one mole of oxidant consumed per mole of reduc-

tant, vide infra. The differential rate law then for reactions

first order in each species may be expressed as follows:

-d[R] -d-0]
- kbs[R][0] (1)
dt dt

where [R] and [0] are the respective concentrations of reluctant
and oxidant and kobs is the rate constant observed at a given

acidity. In some cases the rate was also a function of acidity
as well. Determination of acid dependencies will be discussed with
the specific reactions involved.

Integration of the differential equation above yields, when
the reductant is in excess,

[Ri] x ([R]o [O],)kt [R]o
log ( ) + log ( ) (2)
[]o. x 2.303 Co]o

where the subscript (o) represents initial concentrations and x

represents the concentration consumed at time, t.
The use of Beer's Law (A = cbC, where the terms are absor-

bance, molar extinction coefficient, cell path length in cm, and

concentration in moles per liter, respectively) allows the expres-
sion of equation (2) in terms of experimentally observable para-
meters. If one or more of the reactants and/or products has a







characteristic absorbance at the wave length monitored, equation
(2) may be expressed as

[R]o [R]o
At + ( -- )A ( )A,
[o], C[o],
log
At A

( [R]o [O]o)kt [R]o
+ log ( ) (3)
2.303 [o].

where A is the absorbance observed at the respective times (0, t,
and ). A plot of the left side of equation (3) vs time allows
an observed rate constant to be determined (i.e., slope =

([Rio [O]o)k
2.303
wherever feasible a large excess of reductant ([R]o > 10[0])
was used. Under this condition equation (3) may then be reduced
to the form

kot
log (A A ) = + log (Ao A,) 4)

where kobs = [Rlk. In this case a plot of log (At A) vs t
allows an observed rate constant to be obtained. Use of the
average value of [R]o reduces the inherent error in the approxi-
mation leading to equation (4). The derivation of equations (3)
and (4) can be found elsewhere.10
For the cases where the forward and reverse rate constants
were to be evaluated for an equilibrium reaction of the type

k+ kf
Z +H T P
r







where [P]o = 0, [Z]o = Z, and hydrogen ion was present in large

excess, equation (4) may be utilized by replacing A, with Aq.

Thus

k t
log (A Ae) = + log (A. Aq) (5)

where kbs = kf + kr. The forward and reverse rate constants
were thus derived from


kobs kobs Eeq]
k + (1/K [H+]) 1 + Keq[H] Kq eq [H

Elucidation of this method is discussed elsewhere.3031


Evaluation of Activation Parameters


Consider the simple bimolecular reaction

K* k*
R + 0 (RO)* Products

where species R and 0 form the activated complex (RO)* which then
may proceed to products. Following methods previously outlined,32
K* can be treated as an equilibrium constant and the reaction rate
expressed as

-dER] kT
--] = k [R][O] = k*K*[R][] = K*[R][OI (6)
dt h


and kp = ()K* (7)

where k is the observed rate constant, k is Boltznanns constant,
P








h is Planck's constant, and T is the absolute temperature. Treat-

ment of K* as an equilibrium constant allows the expression

AG* = -RT In K* (8)

to be used. Then

k = L e-AG*/RT = kT_ AS*/R e-IH*/RT ()
pk e e eh(9)

Equation (9) may then be converted into the form of a linear

equation to yield

k
log (y) log ( .) + 2.3R (10)
k 1
A plot of log (-) vsy then yields a value for AH* from the

slope, m:

AH* = -2.3R(m) (11)

The entropy of activation can be obtained from the y-intercept, b,

by using equation (12),


AS* = b- log ( )2.3R














RESULTS


Characterization of Complexes



Spectral and ion exchange properties upon which the charac-

terization of all complexes are partially based are summarized in

Tables I, II and III. Proton magnetic resonance data for the

methyl, methylene and methinyl protons of interest are listed in

Table IV.

2-Mercaptopropionatobis (ethylenediamine) cobalt(II) Perchlorate-
[Co(en)2(OOCCH(CH3)S) ]C10

Characterization of this complex was accomplished by elemen-

tal analysis, infrared, visible, ultraviolet, and proton magnetic

resonance spectroscopy, ion exchange chromotography, acidity

studies and reaction patterns both independently and in comparison

with the previously reported mercaptoacetato complex.0 The basis

for the formulation of the complex as written, bound through car-

boxylate oxygen and mercaptide sulfur, will now be summarized.

Elemental analyses for carbon, hydrogen, nitrogen, sulfur

and cobalt as previously reported in the Experimental section were

in good agreement with the calculated percentages. The infrared
-l
spectrum of the complex exhibited no absorption in the 2,500 cm

area which substantiates the sulfur being deprotonated in the com-

plex as a solid. Intense absorption at 1,630 cm-1 and 1,340 cmn-







TABLE I

Elution Characteristics of Some Cobalt(III) and Chromium(III) Speciesa


Species Eluentb Vl(ml)' V2(ml) de


is-[Co(en)2(C1)21 1 65 40
[Co(en)2(OOCCH(CH3)S)I+ 1 60 30
2 35 20
[Co(en),(oocCH(CH 3)0)r 1 55 30
[Co(en),(CCH(CH 3)OH) 2) 2 100 50
[Co(en)2(CoCCHSCl3)]2 1 325 130
[Co(en)2(OCCH2NH2)2+j 1 420 140
2 125 70
[Co(H20)62+ 1 210 100
2 90 40
[Co(en)3]31 1 1000 -
2 600 -
[Cr(H20)4(ooCCH(GHC3)S)jf 1 45 25
[Cr(H20)4(ooCCH(CH3)0)]f 1 50 25
[Cr (o)5(oOCCH(CH3)SH)]2+ 2 90 40
[Cr(HO2)5(0oCCH(CH3)OH)]2 2 85 40
[Cr(Hgo) (ooCCH2SCH 3)I2 1 315 50
[Cr(HZo)5 (CCH2H]3* 2 550 200
[Cr(H2 )61+ 1 230 100
2 110 45

a30 x 1.25 cm column in Na form. 0.2 mmole used of each ion. bEluents:
(1) 1.0 t_ NaC101; (2) 0.10 M HC104--2.0 M_ NaC10O. CVolume eluent passed
before band starts to elute. dVolume eluent passed for elution of band.
eApproximate volumes. Duplicate runs varied up to 15%.





Spectral Parameters of Some Cobalt(III) Species

Species A(E) M(E) \(E) pH Ref.

[Co(en.(. .CCH(CH 3)1 )] 517(153) 360(350) 282(12,350) 0-7 a
[co(en) (ooccH(CH3)o0) 517(139) 360(150) 7 a
2+
[Co(e: ,', .C1 )OH)]2+ 499(113) 349(123) 1 a
[Co(en)2(OCCCH2SCH 3) 2+ 499(168) 360(250) 280(7,300) 1-6 a
[Co(en)2(OOCCH2NH2)]2+ 487(98) 346(106) 0-7 a
[Co(en) (OOCCH2NH2)]2+ 487(98) 346(107) -- 0-7 2
LCo(er', .. ))(OOCCH(CH3)OH)]2 498(-100) 360(-80) -- 1 a,b
[Co(en)2(COCCH2S)] 518(152) 282(12,600) 0-7 10
[Co(en)2(oOCCHg2) 1 518(132) 360(140) 7 10
[Co(en)2(OOCCHgoH)] + 499(113) 348(122) -- 1 10
[Co(e:r' ( 0)(OOCCH2 OH)]+ 498(i100) 360(-80) -- 1 10
[Co(en)2(OCH2 H2S)T+ 524(139) -- 283(12,600) 7 10
[Co(en) 2(OHcH2CHC2S) 2+ 482(138) 370(283) 282(13,200) 0-7 19
[Co(Nl3)5(ooccH(CH3)OH)]2+ 503(73) 350(62.1) -- 7 9
cis-[Co(en)(o)(OOCCH3)] 512(101) 365(85) -- c
cis-[Co(en)2 (H20)(OOCCH3)]+ 498(100) 360(77.5) ?
cis-[Co(en)2(Cl)(NH2CH2CO0))] 525(80) 365(83) -- 2


aThis work. -rom hydrolysis of [Co(en)2(OOCCH(CH3)OH)]2+. CV.
G. Ortaggi, Inorg. Chem., 6, 2168 (1967).


Carunchio, G. Illuminati, and




TABLE III
Spectral Parameters of Some Chromium(III) Species


Species A(N) \(E) \(E) pH Ref.

[Cr(H0)4(OOCCH(CH3)S)]+ 545(71.2) 440(52.2) 265(5,050) 1-2 a
[Cr(H20)5(OocCH(CH3)SH)]2+ 568(25.0) 411(24.3) -- 1-2 a
ECr(H0) (OCH(CH3 )o0)1+ 548(31.0) 437(38.0) -- 52.8 a
[Cr(H2 )(OOCCH(CH3)O)]+ 570(40.1) 411(45.0) -- 35
[Cr(H20)5(OOCCH(CH3)OH)]'2+ 568(26.8) 413(33.2) -- 1 a
[Cr(H20)5(OOCCH(H )OH)]'2 563(25.7) 409(31.2) 41
[Cr(H20)5 (OOCC2SCH )2+ 567(26.7) 412(25.9) -- 1-2 a
[Cr(H20) 5(OOCCH2SH) 2+ 568(26.0) 411(25.1) -- 1-2 10
[Cr(H20)4(0CCH2NH2) 2+ 555(38.0) 420(41.0) -- 4.5 a
[Cr(HO2)5 (OOCCHNH )]3+ 572(22.0) 412(23.0) -- 3 a
[cr(H20) 4(o CH2S) 548(68.3) 437453.4) 264(5,070) 1-2 10
[Cr(Hz)4(OOCCH20)]+ 548(32.5) 436(38.5) -- 2.8 10
[cr(H o)5(0OCCHO2H)1]2 568(24.5) 411(30.5) -- 1 10
[Cr(H20)5(SH)]2 574(27) 434(43) 259(-7,000) 0-2 34
[Cr(H o)5(OOCCH 3)]2 570(24.4) 410(22.2) -- 7 25
[Cr(H2o)6]3 574(13.4) 408(15.6) -- 7 43

aThis work.







TABLE IV
PMR of Various Protons Adjacent to Chalcogens

Species Medium Methinyla Methyla Methylenea Ref.
Proton Protons Protons

HOOCCH(CH3)SH CDC13 3.58 1.56 -- b
DOOCCH(CH )SD D20 3.65 1.48 -- c
NaOOCCH(CH )SD D20 3.45 1.35 -- c
[Co(en)2(OOCCH(CH3)S)] D20 3.51 1.32 -- c
[Co(en)2(OOCCH(CH )S)1+ D20-CF3COOD 3.51 1.32 -- c
NaOOCCH(CH3)OD D20 4.09 1.30 d
[Co(en)2(OOCCH(CH(3)0) D20 3.92 1.25 -- c
[Co(en)2(OOCCH(CH 3)OD)]+ D20-CF3COOD 4.36 1.32 -- c
CH3SC2CH2COOCH3 D20 -- 2.08 2.55 d
CH2(SCH2COOH)2 CF COOD -- -- A=3.62 d
[Co(en)2(OOCCH2SCH)2+ D20 2.38 3.75
[Co(en)2(0OOCHCH2SC3)] D20-CF3COOD -- 2.38 3.75 c
D2NCH2COOD D2 -- 3.58 d
[Co(en)2(ooCC2H2 H 2)]2 D20 -- 3.58 e
NaOOCCH2SD D20 3.51 d
HOOCCH2SH CDC13 3.31 d
[Co(en)2(OOCCH2S)]+ D20 -- 3.10 10
[Co(en)2(OOCCH2S)]+ D20-CF3COOD -- 3.10 10
NaOOCCH20D D20 -- 3.99 d
DOCC20D D20 4.29 d
[Co(en)2(OOCCH20)] D2 -- 3.90 10
[Co(en)2(OOCCH20D)7]+ D20-CF3COOD 4.26 10
CH3SCH2CH(ND2)COOD D20 4.51 2.22 3.27 d
[Co(en)2(OOCCH(CH2SCH3)NH2)]2+ D20 3.60 2.22 3.08 39

aDownfield from TMS (in ppm). barian Associates Catalog, Vol. II. CThis work.
dSadtler Tables. eD. H. Williams and D. H. Busch, J. Amer. Chem. Soc., 8Z,
4644 (1965).








indicate the presence of coordinated carboxylate.,' These

data are consistent with the ion exchange studies which indicated

a unipositive ion (Table I).

Spectral parameters in the visible and near-ultraviolet for

the compound are as follows: [N(E): 517(152 t 2), 360(340 t 20),

282(1.23 t 0.02 x 10 )]. The error associated with the molar

absorptivity at 360 nm corresponding to the T2 --A, d-d trans-

ition is due to its super position on the tail of the intense band

at 282 nm. This contributes to the rather high value for the molar

absorptivity at 360 nm with intensity borrowing from the higher

energy peak also appearing important. It was necessary to use the

ultraviolet rather than the visible source because the slit width

with the former allowed for better resolution at the 360 n= setting

to define the shoulder clearly.

The near-ultraviolet peak is assigned to sulfur-to-metal

charge transfer due to its large molar absorptivity and the appear-

ance of similar peaks (2 nm) in the spectra of the 2-mercaptoace-

tato, 2-mercaptoethanolato and 2-mercaptoethylamine complexes.

Similar peaks are absent in the spectra of the oxygen analogues.

A fourth absorbance with maximum near 220 nm and molar absorptivity

of 19,000 + 2,000 was noticed for the thiolactato, lactato, mercap-

toacetato, glycolato and glycinato species, but reproducibility is

less than desirable. It is tempting to assign this transition to

carboxylate oxygen-to-metal charge transfer which would then corre-

late the lower energy sulfur-to-metal charge transfer with the

greater ease of oxidation of sulfur relative to oxygen.

As for the mercaptoacetato complex,l0 no change in visible or








near-ultraviolet spectral parameters over the pH range 0-7 lead

to the conclusion that the thiolactato complex is not detectably

basic under the conditions studied. Kinetic data support this

observation, vide infra. The lack of basicity in this range is

comparable to the results observed for the mercaptoacetato cobalt

complex10 and thiolochromium complex.34'35

Proton magnetic resonance spectra of several cobalt(III)-

chalcoganide complexes previously have shown that the resonance of

the methylene protons contiguous to the coordinated chalcogenide is

shifted upfield by 0.2 o.7 ppa from the neutral free ligands and

that deuteronation of coordinated alkoxide shifts the methylene

resonance to near that of the free ligand.10,19.36 It has also been

demonstrated that the resonance of methylene protons adjacent to

coordinated sulfur remains shifted equally upfield from the free

ligand in both neutral and acidic media.10'19

From these results it was expected that the methinyl proton

on the propionate skeleton would exhibit similar behavior. The

resonance of the methinyl proton adjacent to coordinated sulfur in

the 2-meroaptopropionate skeleton is shifted upfield from the neu-

tral free ligand resonance but by the smallest amount yet observed,

0.14 ppm. Such a small shift seems inadequate to diagnose sulfur

coordination except in combination with other supportive data such

as that presented here. The methyl resonance is similarly shifted

upfield by only a small amount, 0.16 ppm. The lack of response in

the methyl resonance to acidification supports the contention that

sulfur is not protonated over the acid range studied. The reacti-

vity patterns and product studies described below provide further

confirmation of the formulation of the complex as described.








Lactatobis(ethylenediamine)cobalt(III) Perchlorate--{Co(en)2(OOCCH-

(CH )0)]C104

Similar techniques were employed for the characterization of

this complex. Elemental analyses for carbon, hydrogen, nitrogen

and cobalt were in good agreement with the calculated values for

the complex as a unipositive ion coordinated through both carboxy-

late and alkoxide oxygen atoms (Experimental section). Further,

the complex eluted as a +1 ion with neutral eluent and as a +2 ion

with an eluent of pH = 1 using the previously described ion exchange

methods (Table I).

Infrared spectral data confirm the presence of coordinated

carboxylate which exhibits maxima at 1,630 cm- and 1,350 cm-.

Determination of the presence or absence of coordinated alkoxide

by examination of the 0-H stretching region was rendered impossible

due to broad N-H and C-H stretching mode absorption originating from

the ethylenediamine ligands in the 3,500 cm-1 and 3,000 cm-1

region.

Spectral parameters in the visible and near-ultraviolet re-

gions are as follows: [X(E): 517(138, 360(153) in neutral water

and 499(113), 349(123) in 0.10 M HC0 4]. An absorption near 220 nm

with unreliable reproducibility in molar absorptivity was also

present.

Determination of the Ka of the complex was necessary due to

its inclusion in the rate expression at low pH, vide infra. A value

was obtained using three methods of determination which gave good

agreement. A pKa of 3.36 at 1.0 M ionic strength (LiC104) was

obtained by spectral methods.25 Direct electrometric titration








with 0.100 N HC1 of one millimole of complex in 100 ml of water

gave pKa = 3.33. A titration with 0.100 N NaOH of one milliiole

dissolved in 25.0 ml 0.100 N HC1 (providing a known excess)

yielded pKa = 3.43 ( = 0.10 M). A value of 3.37 0.06 was

adopted as the pKa of the complex. It can be seen that there is

little, if any, variation with ionic strength as has been found

elsewhere." The value of 3.37 is in reasonable comparison with

those for the analogous glycollato complex (pKa = 3.3 t 0.3) and

the 2-aminoethanol complex (pKa = 3.59).37 The [Co(en)(OOCCH(CH3)-

OH)0 ion is more acidic by a factor of 1027 and 102.2 over the

corresponding cis-diaquo and [Co(NH3)5(HOCH3)i2 complexes with

pKa values of 6.1 and 5.58, respectively.38'35
The pmr spectrum of this complex reveals a resonance for the

methyl group adjacent to alkoxide oxygen which is only slightly

shifted relative to the free ligand monoanion in both neutral and

acidic media (Table IV). In contrast, the methinyl proton reso-

nance is shifted upfield by 0.17 ppm from the free ligand mono-

anion. The comparisons evident in Table IV-suggest that the shift

from the free neutral ligand will be comparable to that observed

for the glycollate complex. Together with the observation that

deuteronation effects a 0.44 ppm shift downfield compared to a

0.36 ppm shift for the glycollate complex, this appears to vali-

date the resonance shift criterion for alkoxide coordination.

The diminished sensitivity of the 2-mercaptopropionate methinyl

resonance to mercaptide coordination is not understood but may

arise from steric effects.








The par spectra, the acidity of the complex, failure to

incorporate water on recrystallization and the striking similarity

in the visible spectra (both in acidic and neutral media) to the

analogous glycollato complex previously characterized and reported

provide evidence that the alcohol function is coordinated to the

metal center in solution. Reactivity patterns and product studies

described below are consistent with this conclusion. Further,

prolonged exposure of the complex to 0.10 M HC104 effects a first-
-7 -1
order reaction (k = 1.9 x 10 sec ) to what is assumed to be

s-[Co(en)2(01i)(OOCCH(CH 3)H)]2+ on the basis of spectral compar-

ison to the cis-acetato-aquo and cis-glycollato-aquo analogues

(Table II).

2-Methylthioacetatobis(ethylenediamine)cobalt(III) Perchlorate--
LCo(en)2(OOCCH2SCH3 ) ]( C10)2

This compound was characterized utilizing the methods employ-

ed for the previous compounds. Elemental analyses for carbon,

hydrogen, nitrogen and sulfur were found to be in good agreement

with the calculated values (Experimental section). The finding

that analyses for hydrogen and sulfur were very slightly lower and

higher, respectively, than the calculated values weighs heavily in

favor of coordination through carboxylate oxygen and thioether

sulfur since the principal probable impurity, that with water

coordinated in place of the thioether function, would reverse the

deviations. Ion exchange characteristics for the complex were that

of a +2 ion (Table I).

Infrared spectral data confirm the presence of coordinated

carboxylate via strong absorptions at 1,630 cm-1 and 1,350 cm ,








which are characteristic of this mode of coordination. 6 The

definition of sharp absorbances at 3,470 cm- and 3,400 cm-1 char-

acteristic of N-H stretching modes argues against presence of

water in the solid since presence of the latter usually obscures

the N-H stretches.3

Spectral parameters in the visible and near-ultraviolet for

the compound are as follows: ((): 499(168), 360(250), 280

(7,300)]. As previously noted for the 2-mercaptopropionatocobalt-
(III) complex, the absorption at 280 nm is taken empirically to be
characteristic of metal-sulfur coordination. This transition

energy is attributable to alternative formulation only with exces-

sive difficulty. In the pH range 1-6 there was no change in the

spectral observations as would be expected for a weakly basic tri-

valent sulfur atom. For the experimental conditions used in this

study the complex was found to be stable in solution for at least

four hours, a period much longer than that employed in the kinetic

measurements.

The advantage of using as a structural probe, the reaction of

a species with a reagent whose reactivity patterns are thoroughly

characterized, is demonstrated convincingly in this case. Reactions

of the complex with [Cr(H20)6J proceeds in two or three observable

steps to yield an isolable carboxylatopentaaquochromium(BII) product

with the rate of the first step being independent of acid concentra-

tion, video infra. This behavior cannot be reconciled with that

expected for the only apparent alternative formulation, [(en)2Co(H20)-

(00CCH2SCHj3) which should be reduced in one step to the carbox-

ylatopentaaquochromium(III) product, or with an inverse acid







concentration dependence to yield [Cr(H20)6]3+, [Cr(H20)5(ooC-

H2SCH 3)]2 or [Cr(H20)5(CH3SC2COOH)] The last complex is, in
fact, believed to be the initial product which can convert to the

carboxylatopentaaquochromium(III) product in two steps, chelation

by carboxylate followed by dechelation at sulfur. However, its

genesis by a rapid, acid independent reaction cannot reasonably

be attributed to reaction with the alternative complex. Its

generation can be rationalized in terms of the formulation pre-

sented, vide infra.

The proton magnetic resonance spectrum of the complex appears

to provide corroborative support for the thioether function being

coordinated. The methyl resonance undergoes a shift of 0.30 ppm

downfield, relative to the esterified methylthiopropionic acid,

while the methyl resonance for the complex of S-methylcysteine in

which the thioether is not coordinated39 remains unshifted relative

to the free ligand value. Further, the methylene resonance is

shifted downfield by 0.13 ppm for this complex, relative to a

comparable ligand, whereas for the pendant thioether complex, the

shift is 0.2 ppm upfield. Thus, if the complex is regarded as

being derived from the mercaptoacetate complex by a methyl car-

bonium ion substitution on coordinated sulfur, the effect on the

methyl and methylene resonances are similar to but larger than

that arising from deuteronation of coordinated alkoxide, vide infra

Admittedly, this represents an entirely empirical approach to the

interpretation of pmr shifts on coordination which must be regar-

ded as tentative and subject to reevaluation as evidence accumu-

lates. Fortunately, the assignment of thioether as an active







donor function in this complex rests as well on the broader
evidence adduced previously.
Glycinatobis(ethylenediamine)cobalt(III) Perchlorate--Co(en)2(00-

CCH2h2)(C104)2

This compound was characterized by elemental analysis, visible
and infrared spectroscopy, and ion exchange chromatography. Elemen-
tal analyses for carbon, hydrogen, nitrogen and cobalt were in good
agreement with the calculated values (Experimental section). Ion
exchange behavior identified the ion as a +2 species with both
neutral and 0.10 14 acid eluent (Table I).
Visible spectral parameters for the complex in water and
0.10 M acid are as follows: [A\(): 487(98), 346(106)]. The
infrared spectrum exhibited intense peaks at 1,640 cm-1 and 1,340
cm-1 characteristic of coordinated carboxylate.'1633 The spectral
parameters are in good agreement with the previously prepared and
reported chloride salt of the complex, [Co(en)2((OCCH0 2)JCl2.6
2-Mercaptopropionatopentaaquochromium(III) Ion (Mercaptide-Bound)-
[Cr(H20) (SCH(C 3)COOH)]"

The reaction of aqueous chromium(II) with [Co(en)2(OOCCH-
(CH3)S)]+ is extremely rapid and results in > 90% incorporation of
the mercaptopropionate in the chromium(III) product coordination
sphere, vide infra. On the basis of these observations and the
high sterit improbability of a doubly bridged mechanism utilizing
both carboxylate oxygen and mercaptide sulfur evident in models
lead to the formulation of the first product of the oxidation-
reduction as LCr(H20)5(U H(CH3)COOH)]2+. When the reaction of the

cobalt(III) complex and chromium(II) ion is performed with a slight







excess of oxidant a fleeting intermediate is observed which under-
-2 -l -1t
goes a rapid subsequent reaction (kos = 3.29 x 102 M- sec- at

H+ ] = 0.90 M, see kinetic description below) to produce a species
spectrally identical to the chelated 2-mercaptopropionatotetra-

aquochromium(III) ion, a sufficiently long-lived species to be

partially characterized in solution and the only isolable product

of the reaction of the cobalt(III) complex in the presence of the

excess chromium(II) ion.

The sulfur-bound monodentate intermediate is not sufficiently

long-lived to determine its spectral parameters, but the relatively

small change in absorbance observed on conversion to the chelated

product suggests an anomalously high molar absorptivity for the

intermediate. Since the intermediate, as formulated, is comparable

to the [Cr(H20)5SH]2+ species (see Table III), the spectral obser-

vations are then reasonable and lend credence to the nature of the

short-lived species as proposed.

2-Mercaptopropionatotetraaquochromium(III) Ion-[Cr(H20)4(OOCCH-

(CH )S)]

The isolable product obtained from the previously discussed

species or from reaction of the cobalt(III) complex with excess

chromium(II) is so formulated on the basis of ion-exchange behav-

ior and spectral parameters observed from the products of the

reaction of [Co(en)2(OOCCH(CH3)S)]+ and Cr(II), a representative

example of which follows.

A 210 ml reaction mixture initially 5.0 x 10-3 L in Co(III),

5.1 x 10-3 M in Cr(II) and 0.020 M in H was exposed to the air

after a five-minute reaction period and charged onto a 30 cm








(1.2 cm diameter) column of Biorad-purified Dowex 50 w-x2, 200,

400 mesh, cation exchange resin in the lithium form. After a

charging and washing time of about thirty minutes a +1 chromium

species was eluted with 0.25 M LiCI04. The band began to elute

after passing about 50 ml of eluent and was collected in about

35 ml. Volume was adjusted to 100 ml with eluent serving as

diluent. Aliquots of this solution were then used for further

study with appropriate additions of standard HC104 and solid

LiC104 to maintain the desired pH and ionic strength. Separate

runs provided 90-95% recovery of the 2-mercaptopropionate-contain-

ing chromium(III) ion, on a 1:1 mole basis relative to the amount

of cobalt(III) complex used initially, based on the previously

described method for chromium determination, vide supra.

That this product is the chelated 2-mercaptopropionate com-

plex is confirmed by the +1 charge and its spectral parameters

C[A(): 545(71.2 1.0), 440(52.2 1.0), 265(5040 150)] (the
uncertainties in the molar absorptivities are due to subsequent

reaction of the species, vide infra). These values compare

favorably with those of mercaptoacetatotetraaquochromium(III) ion

and thiolopentaaquochromium(III) ion (Table III). In both of the

latter species the high energy d-d transition (440 nm) and the

near-ultraviolet absorbance (265 nm, provisionally assigned to

sulfur-to-metal charge transfer, vide supra, strongly suggest

mercaptide coordination. The fact that the low energy d-d transi-

tion (545 nm) occurs for both acid complexes and is blue-shifted

from the values for the thiolo and aquo chromium species indicates

carboxylate coordination (Table III). Similar observations have








been reported for several carboxylate-bound chromium(III)

species.35 The high molar absorptivities observed suggest

chelation, which would lower the overall symmetry. Examination

of the molar absorptivities of several cis-aquo carboxylate com-

plexes of both cobalt(III) and chromium(III) further substan-

tiate this explanation (Table II and III). Occurence of the

sulfur-to-metal charge transfer absorbance some 20 nm higher in

energy for chromium(III) complexes than for cobalt(III) complexes

lends credence to the assignation of the transition to the charge

transfer in view of the decreased tendency of chromium(IllI) to be

reduced relative to that of cobalt(III), (Ered (3+/2+): +1.84 v

(Co3+/Co2+), -0.41 v (Cr+/Cr2+)).40

2-Mercaptopropionatopentaaquochromium(III) Ion (Carboxylate-
Bound)--Cr(H20)5 (OOCCH(CH)SH)]2+

Isolation in solution of this species from the direct

reaction of the cobalt(III) complex and chromium(II) is not possible

due to presence of the cobalt(II) ion produced. Elution character-

istics of the product +2 ions are nearly the same, resulting in an

ineffectual separation. However, isolation of the previously

described chromium(III) chelate complex followed by subsequent reac-

tion in various acid solutions affords a route to an equilibrium

mixture of the chelate and pendant (carboxylate-bound) chromium(III)

ions. This mixture can then be separated by ion exchange. Use of

the ionic strength in the range 0.25 M to 0.50 M in HC104--LiCD4

for the equilibrium reaction results in the chelate fraction not

being retained on the column while the +2 monodentate form remains.

Elution with 1.0 M LiClO4 allows isolation of the monodentate form.







Because of subsequent reaction of the monodentate carboxylate

ion to hexaaquochromium(III) it is not possible to recover total-
ly the initial amount of chromium as either chelate or monodentate.
The monodentate product was characterized by its +2 charge
and the similarity of its visible spectrum [A(c): 568(25.0 t 1.0),
411(24.5 1 1.0)] to previously prepared carboxylatochromium(III)

ions.35 Again a high energy peak at about 210 nm was present, but
again of varying intensity (e = 19,000 t 2,000).
Upon isolation, solutions of the monodentate carboxylato
product can be observed to undergo subsequent reaction spectro-
photometrically. Ion exchange separation of an equilibrium mix-
ture after several days results in three separable fractions. The
first fraction elutes as a +1 ion and is spectrally identical to
the original [Cr(H20)4(OOCCH(CH3)S)]1 ion whereas the second frac-
tion is the monodentate species, the +2 ion [Cr(H20)5(OOCCH(CH 3)-
SH)f The presence of the chelate form shows that dechelation
of the chromium-sulfur bond is reversible. The third fraction was
identified as [Cr(H20)6]3,
Lactatopentaaquochromium(III) Ion (Alkoxide-Bound)--[Cr(H20) (OCH-

(CH )COOH)]2+

That this species is the first product formed in the reaction
between aqueous chromium(II) ion and [Co(en)2(OCCH(CH3)OH)]2 at
pH = 3.4 (where the alcohol function is substantially deprotonated)

or of the path inversely proportionalto hydrogen ion concentration
at low pH, vide infra, follows logically in comparison with the
sulfur analogue. The ion has not been independently identified,
however.







Lactatopentaaquochromium(III) Ion (Carboxylate-Bound)--[Cr(H20)-
(ooccH(CHOl. i )12+

From the reaction of [Co(en)2(OOCCH(CH )OH)+ (1.0 x 10-

I), with chromium(II) (1.1 x 10-2 M) in 0.100 M HCIO4, a diposi-
tive chromium(III) product, characterized as the monodentate lacta-
to complex, can be isolated using ion exchange techniques. The
visible spectrum in 0.10 M HC104 had the following parameters:

[>(E): 568(26.8 t 1.0), 412(33.2 t 1.0)]. The molar absorptivi-
ties are slightly higher than previously reported values of 25.7

and 31.2 for the respective peaks, but it should be noted that
the complex was not isolated in the previous work and values are
based on 35% complexation in a solution of lactate and chromium(III)
ion. It should be noted further that the peak ratios reported pre-
viously (EA/E~2 = 0.82) do agree reasonably well with values ob-
served in this work, (EA/E 2 = 0.81).
When the reaction of [Co(en)2(OOCCH(CH3)OH)1 was performed
at an initial acidity of 4.0 x 10- M HC104, two chromium(IIl)-con-
taining complexes of the lactate were eluted from the cation ex-
change column in the approximate molar ratio of 2:1. The first and
major fraction was eluted with neutral eluent as a pink +1 ion
described below. The second fraction eluted with the same neutral
eluent as a +2 ion that changed over short but observable periods
from a blue color on the column to a pink color in solution with a
spectrum identical to that of the first fraction (pH = 2.9). The
blue species may be the alkoxide-bound monodentated complex but its
fleeting existence makes this characterization tentative. Acidifi-
cation of both fraction to pi = 1 gave the spectrum of the monoden-







tate species within the time span necessary for the experiment
(5-10 min).

From the fact that the +2 and +1 ions are readily inter-
convertible at a measurable rate, vide infra, it can be surmised
that the +1 ion is not the [Cr(H20)4(0H)(ooCCH(c3)OH)] complex
for two reasons: (1) The pKa of the carboxylate-bound +2 lactato
complex would be expected to be near that of the acetato analogue,

[Cr(H20)5(OOCCH3)]2+, that is, pKa = 4.5.25 (2) If the hydroxy
complex were important, its formation would be expected to be
diffusion controlled.
Lactatotetraaquochromium(III) Ion--[Cr(HO20)4(OOCCH(CH?)0)1

This chelated species was concluded to be the major product
observed ("65%) from the reaction of [Co(en)2(OOCcH(CH3)0)] and
chromium(II) at pH = 3.4. Cation exchange separation characterized
the species as a +1 ion which had spectral parameters of [A(W):

548(31.2), 437(38.1)] at pH = 2.8. Subjecting the monodentate
carboxylato species to pH = 2.8 resulted in a change producing sim-
ilar spectral parameters based on total chromium(III), [>(e):

548(31 2), 437(37 t 2)], within the five minutes necessary for
manipulation.
Examination of the spectral parameters of the proposed che-
late complex and comparison to the analogous mercaptopropionate
complex, which appears more solidly formulated, supports the chelate
form for the +1 complex. The energies for the d-d transitions are
essentially identical for both the alkoxide and mercaptide complexes
of chromium(III) (Table III) as is found to be the case for the
cobalt(III) complexes (Table II). The red-shifting of the high








energy band and blue-shifting of the low energy band relative

to chromium(III) occur in the regions expected for coordination

through alkoxide and carboxylate but not expected for coordina-

tion of either function alone. Finally, the observed molar ab-

sorptivity would be anomalously high for chromium(III) complexes

bound through carboxylate alone in comparison to similar com-

plexes.35

Metthylthioacetatopentaaquochroium(III) Ion (Thioether-Bound)-
[Cr(-20) 5(S(CH )CH2CO0H) )

By comparison to the mercaptide complex, this species is

expected to be the first product formed in the reaction between

aqueous chromium(II) ion and [Co(en)2(0CCH2SCH 3)j Since the

ion has not been independently identified and its formulation is

based on kinetic results, discussion of it will be deferred to

the appropriate section.

Methylthioacetatotetraaquochromium(III) Ion--[Cr(H20)4(OOCCH2SC-



Formulation of this ion as the product of a subsequent

reaction of the thioether-bound monodentate species described above

is based on kinetic results and will therefore be discussed in a

later section.

Methylthioacetatopentaaquochromium(III) Ion (Carboxylate-Bound)-
[Cr(H20) (OOCCH2SCH3 )]2+

This species can be isolated in solution in >90% yield using

cation exchange techniques from the reaction of [Co(en)2(OOCCH2SC-

H3 )2+ with chromium(II) ion with the latter in deficiency, equi-

valency or excess. The species is formulated as a +2 ion on the







basis of ion exchange elution behavior (Table II). The visible
spectrum in 0.50 M LiC104, pH = 3.4, exhibited the following para-
meters: [EA(): 567(26.7), 412(25.9)]. The visible spectrum was
found to be quite similar to that of the [Cr(H20)5(ooccaCSH)' and
other carboxylato-chromium(III) species (Table III), thus confirming
formulation of the ion as described. The +2 ion is the only
observable product of the reaction of [Co(en)2(OOCCH2SCH3)1+
and Cr(II) at acid concentrations in the range 0.10 M to 0.10 g.
Glycinatopentaaquochromium(III) Ion (Carboxylate-Bound)--[Cr(H20)5-

This species can be isolated in solution in > 90% yield using
cation exchange techniques from the reaction of [Co(en)2(OOCCH2-

S2)y2+ with chromium(II) with the latter in deficiency, equiva-
lency or excess. The blue product is formulated as a +3 ion on
the basis of ion exchange elution behavior (Table I). The visible
spectrum in 1.0 M LiC104, pH = 3.5 exhibited the following para-
meters: [M(): 573(22.0 1.0), 412(23.0 1.6)]. Although the
peak positions are quite similar to [Cr(H20)6]3+, the molar
absorptivities eliminate the possibility of this complex as a
product (Table III). Spectral comparison to other carboxylato-
chromium(III) species coafira formulation of the ion as described.
The +3 ion is the only observable product of the reaction of
[Co(on)2(00CC1NI22)]2+ and Cr(II) at acid concentrations in the
range 0.1 M to 1.0 x 104 M.
Glycinatotetraaquochromium(II) Ion--[Cr(H20)4(00CCH2NH2) ]+

Upon adjustment of the pji to 4.5 of a solution of the
glycinato(carboxylate-bound)-chromium(III) species described








above, the solution color changed rapidly from blue to green, but

the maxima of the visible absorptions remained essentially constant.

Letting the solution equilibrate for several days resulted in the

solution acquiring a red-violet hue. The higher energy absorption

remained very nearly at its previous position, but the low energy

absorption shifted to 560 5 nm. On passing the equilibrated

solution through a cation exchange column in the lithium form, it

was noticed that a faintly colored fraction was not retained on

the column, probably due to the high ionic strength (1.0 M) of

the charging solution. Collection of this fraction and spectral

analysis yielded the following: [D(A): 555(38.0 5), 420(41.0

t 6)]. From the high molar absorptivities compared to carboxy-

late-bound chromium(III) species but relatively normal for

chelate species (Table III), the spectral shift of the low energy

band to higher energy and the fact that the ion was not retained

on the cation exchange column, it is tentatively concluded that

the species is the chelated [Cr(H20)4(OOCCH,2H2)2+ ion.


Reactions of Chromium(II) with the Cobalt(III) Complexes


2-Mercaptopropionatobis(ethylenediamine)cobalt(III) Ion--
[Co(en)2(OOCCH(CH2)S)]1

The stoichiometry of this reaction was determined by ion

exchange separation of the reaction products followed by chromium

analysis of the appropriate fraction. Several reactions were

carried out having reluctant and oxidant alternatively in excess.

A solution of chromium(II) was injected into solutions of the








complex with the usual exclusion of air. Within fifteen minutes

the reaction mixture was exposed to the air and charged onto a

suitably prepared column. In each case the amount of chromium(III)

product elutable as a +1 ion closely approximated that of the

deficient reactant (Table V). This is indicative that the

oxidant and reductant react in equimolar amounts. This first

isolable product was characterized as the chelated chromium(III)

product, [Cr(H20)(0OCCH(CH3)S), vide sura. The fact that

the ligand is 100% incorporated in the product at varying reac-

tant ratios is taken to establish the reaction as inner-sphere
42
in nature.42 As previously discussed, the chelated +1 ion was

concluded to be a secondary product of the initial oxidation-

reduction reaction, the result of the ring closure of the primary

monodentate product, [Cr(Ho0)5(S(CH3)CHcooH) The only other

chromium-containing fraction was obtained from reactions with

reductant in excess and was characterized spectrally as the

air-oxidized dimer of chromium(III).43 Spectra of the product

solutions corresponded within 4% to those calculated based on a

1:1 stoichiometry and summing the contributions of each species

remaining.

The rate of the Cr(II)-[Co(en)2(OOCCH(CH )S)J' reaction

was found to be measurable on the stopped-flow instrument. Since

the second-order rate constant was >105 (M1 sec-1) it was

necessary to follow the disappearance of the sulfur-to-metal

charge transfer peak in the near-ultraviolet region. Here advan-

tage could be taken of the high (-12,000) molar absorptivity which

allowed solutions of extremely low concentration to be used.








TABLE V

Stoichiometries of the Reactions of
Cobalt(III) Complexes with Chromium(II)


Co(III)L Cr(II) [H+] Cr(III)L+ Cr(III)L2+ Cr(III)L-3
mole mole M mole mmole mmole


L = 2-Mercaptopropionate

0.241 0.450 0.100 .214 -
1.00 1.00 0.0100 .933 -
0.245 0.150 0.100 .141 -

L = Lactate
0.251 0.500 0.100 -- .240
0.244 0.250 0.100 -- .232
0.249 0.150 0.100 -- .141
0.050a 0.050 4.0 x 10" .004? .0023 -

L = Methylthioacetate
0.102 0.175 0.100 .089 --
0.100 0.094 0.100 -- .086
0.092 0.065 0.100 .065 -

L = Glycinate
0.202 0.400 0.100 -- .190
0.205 0.208 0.100 -- .190
0.143 0.064 0.100 .060

aReaction incomplete after 15 hours.








In a typical experiment, solutions of cobalt(III) complex at
3.75 x 10 M and chromium(II) at 5.4 x 10- M each at H =

1.0 x 10-2 M and with ionic strength maintained at 0.10 M_ with

LiC104 were deaerated while thermostatted, then transferred to the
drive syringes of the stopped-flow instrument as previously

described. Use of these concentrations allows the data obtained
to be treated as a pseudo-first-order reaction. Thus, a plot of

log (At A, ) vs time permitted calculation of an observed
rate constant which then could be converted to the second-order
rate constant. For the concentrations specified, the reaction

was complete in 20 msec. To ascertain the dependence, or lack

thereof, on acid concentration and order of the reaction with
respect to chromium(II) ion, each was varied independently over

a ten-fold range while other variables remained constant. It was

found that the reaction is independent of the acid concentration
in the range [H+] = 0.10 M to [H+] = 0.010 M and first-order
in chromium(II) ion in the range [Cr(II)] = 2.15 x 10-3 M to [Cr-

(II)] = 2.15 x 104 M (Table VI). By virtue of the linearity of the
log (A A, ) vs t plots over at least three half-lives (>90%
reaction) it was concluded that the reaction is also first order

in oxidant. The second-order rate constant for the reaction of

[Co(en)2(OOCCH(CH3)S)] with chromium(II) was found to be (1.55 t
0.25) x 105 M1 sece".
L_.._.'.1. .* ._i, _-_1. n.: -' ii ri_ ,-r.ht ItIlI) Ionr--Cc, n)- !CCH 'i-


The stoichiometry of this reaction was found to be approxi-
mately equimolar in each reactant at [H"1 = 0.010 n (Table V) by








TABLE VI

Acid and Chromium(II) Dependencies of the Reactions of Some
Cobalt(III) Complexes with Chromium(II)


[Co(III)L]n [Cr(II)] [H+3 kob

M x 105 M x 103 1 sec-1


L = 2-Mercaptopropionatea


1.05
1.05
2.15
0.54
0.21


0.100
0.010
0.010
0.010
0.010


L = Methylthioacetateb

47.0 0.100
47.0 0.050
47.0 0.010
98.0 0.100
9.4 0.100

b
L= Glycinate

8.40 0.100
8.40 0.010
98.0 0.100


1.60 x 105
1.55 x 105
1.61 x 105
1.32 x 105
1.46 x 105


2.22
2.26
2.05


af = 0.100 M (LiC1O4--HC104). bA = 1.00 M (LiC14--HCI04).


0.40
0.40
0.40
0.40
0.40


48.0
48.0
48.0








methods analogous to those used in the preceding section. The

reaction was found to proceed by two different pathways, vide

infra, but usually only one chromium-containing product was isol-

able from a given reaction. It has been found that the initial
chromium-containing product of the reaction exists in two rapidly

interconvertible forms, vide infra.

By methods similar to those described for the mercaptide
analogue it was determined that the oxidation-reduction reaction

was first order in each reactant by pseudo-first-order techniques
(Table 1IT). It was possible to follow the disappearance of the

protonated form of the cobalt(III) complex by spectrophotometric

monitoring of its absorption maximum at 499 an on the stopped-
flow instrument. Again linearity over >90% of the reaction was
observed for the kinetic plots. The second-order rate constant

for this complex was found, however, to vary inversely with

acidity over the range 1.0 M [H+1] L 6.7 x 10-3 N (Table VII).
A plot of kobs vs [H ]- was linear (Figure 1) and yielded the
expression for the rate constant as


7.35 + 0.023 [H'-1
k = (M and see)
1 + K .+ ]-1

This observation can be understood from a rigorous solution
of the differential rate equation for this reaction10

-d[Co(III )]total
dt = kobs[Co(III)]total (13)
where Ka is the acid dissociation constant of [Co(en)(OOCCH(CH3)-
OH)]2 and k1 and k2 are defined by












TABLE VII
Acid and Chromium(II) Dependence of the Reduction of
[co(en)2(OOcCHi(cHOH)]o)2 by Chromium(II)


[Co(III)] [Cr(II)] [H"] kobs
x 10M x 103 M sec-1


2.00 68.0 .0100 9.15
2.00 5.50 .0100 9.62
2.00 5.50 1.000 7.21
2.00 5.50 .1020 7.50
2.00 5.50 .0507 8.10
2.00 5.50 .0167 8.70
2.00 5.00 .0067 10.77
.10 1.00 8.00 x 10-5 47*


Approximate value; variations between separate reactions > 20%.













































Figure 1. Acid dependence of the reduction of [Co(en)2(OOCCH(CH3)OH)
by chromium(II).








k2
[Co(en)2(OOCCH(CH3 )o)] + Cr," -4 products


[Co(en)2(OOC(CH3)O)3" + Cr2+* products
-2
From graphical analysis kKa is found to be 2.3 x 10 and Ka

determined to be 4.4 x 10 vide supra, leading to the calcula-

tion of k = 52. Several experiments were performed at [H + =

8.0 x 10-5 where the ratio of deprotonated to protonated forms

is -5 with the observed rate being ~47. The large uncertainty in

rates (>20) encountered at this acidity level, probably due to

consumption of protons by the free ethylenediamines released in

the reaction, renders the values approximate. Their usefulness

is limited to the observation that as the concentration of the

deprotonated form is increased the observed rate constant approach-

es the predicted value.

From the observed rate law it is obvious that at 0.100 J

acid the immediate product should arise almost exclusively via

the k, path. As the acid concentration decreases, however, the

product of the second path should appear and become dominant at

extremely low acid concentration. At 0.100 M acid [Cr(H20)5(0O-

CCH(CH )OH)]2+ is the only isolable product. By using a large

volume and extremely low concentrations of reactants over a long
period of time, it is possible to observe both the +1 and +2

products in a molar ratio of 2:1 at a pH = 1.0 x 104 .
The fact that both products are isolable from the reaction

is not sufficient to confirm their genesis by two separate

mechanistic pathways since both products are subject to subsequent







reactions over the time period needed for the reaction and isola-

tion period. In fact, their rapid interconversion as a function

of acidity, vide infra, over a time period much shorter than that

required for isolation suggests a thermodynamic rather than a

kinetic distribution of products. Further consideration of the

mechanism is postponed until after the rates of substitution are

presented.

2-Methylthioacetatobis(ethylenediamine)cobalt(III) Ion--[Co(en)2-

(OOCCH2SCH )]2+

Using the ion exchange techniques previously detailed, the

stoichiometry of the reaction between chromium(II) and this ion was

found to be equimolar (Table V). The only isolable reaction product

was characterized as a +2 ion (Table I) and assigned the structure

[Cr(HO0) .-oC*liL. -.JH,+ based on its spectrum (Table III). As
will be discussed in the appropriate section, this species was con-

cluded to be a tertiary product of the initial oxidation-reduction

reaction, the result of ring closure and subsequent dechelation of

the primary monodentate product, [Cr(H20)5(CHSCH2COOH)+ .
The rate of the electron transfer reaction between this co-

balt(III) complex and chromium(II) was found to be conveniently

measurable on the stopped-flow instrument. The decrease in

absorbance at 499 nm was followed spectrophotometrically under

pseudo-first-order conditions allowing data to be conveniently

evaluated, vide supra. A typical reaction was under conditions

similar to or the same as the following: [Co(III)] = 1.3 x 10

M, [Cr(I)] = 4.70 x 10-2 M, [H+] = 0.100 M and ionic strength

maintained at 1.00 M (HC10--LiC104).







As for the previously described cobalt(III) complexes, acid

concentration and chromium(II) concentration were varied inde-

pendently over at least a ten-fold range to evaluate the dependence

of the reaction on these two variables. Linearity of log (At- A,)

vs t plots over at least 90% reaction for the ranges [Hf = 0.100 _

to 0.0100 M and [Cr(II)] = 9.8 x 102 M to 9.4 x 10-3 M with no

significant deviation in observed rate constant leads to the

conclusion that the electron transfer reaction is first order in

both oxidant and reductant and independent of the acidity within
the ranges specified. The calculated second-order rate constant

was found to be (267 18) M1 sec-1 (Table VIII).

Glycinatobis(ethylenediamine)cobalt(III) Ion--[Co(en)2(OOCCH2

NH2) 2+

The stoichiometry of this reaction was determined by ion

exchange separation of the reaction products as described previous-

ly. The elution characteristics of the product are those of a +3
ion (Table I). Due to the longer reaction time relative to the

three previously discussed complexes, the possibility of loss of the

primary product must be considered. Still, recoverable amounts of
the only isolable product, characterized as [Cr(H20)5(OOCCH2-

NH )]3, vide supra, approach closely those expected for equimolar
stoichiometry over the range of reactant concentrations considered
(Table V).
The rate of the [Co(en)2(OOCCH2N.N12)]2+-Cr(II) reaction was

found to be conveniently measurable using the all-glass apparatus
with the Cary 14 instrument as described in the Experimental sec-

tion. A typical reaction was with solutions 4.8 x 104 M in the




TABLE VIII
Rate Constants for Reactions of Co(III) Complexes with Chromium(II)

Species [Co(III)] [Cr(III)J [H+f T kob
M x 105 M x 103 M C M-1 sec-

[Co(en)2(OOCCH(CH3)S)1+ a .40 1.05 0.010 25.0 (1.53 .06) x 105
.40 1.05 0.010 30.1 (1.56 + .13) x 105
.40 1.05 0.010 39.1 (1.72 .07) x 105
.40 1.05 0.010 44.6 (1.82 .25) x 105
[Co(en)2(OOCCH(CH3)OH)+2 b 43.2 8.40 1.00 14.0 5.96 .10
43.2 8.40 1.00 25.2 8.34 .65
43.2 8.40 1.00 34.0 10.72 .18
[Co(en)2(OOCCH2SCH3)]2+ b 13.6 47.0 0.100 19.9 211 1 11
13.6 47.0 04100 25.0 267 t 18
13.6 47.0 0.100 30.2 356 10
13.6 47.0 0.100 36.7 489 10
[Co(en)2(OOCCH2NH)]2 1 b 50.0 8.50 0.100 13.5 1.15 .06
50.0 8.50 0.100 25.0 2.22 + .12
50.0 8.50 0.100 34.1 3.52 .13

aA = 0.100 M (LiC104--HC104). b = 1.00 M (LiC104--HC104).








cobalt complex and 8.4 x 10-3 M in Cr(II) at [H+] = 1.00 x 10-2

M (ionic strength maintained at 1.00 M with LiC104). These were

transferred anaerobically to the glass mixing apparatus ihich was

thermostatted prior to reaction. Again pseudo-first-order condi-

tions were maintained and the data treated appropriately.

In order to determine the acid dependence the [H ] was varied

from 0.100 M to 0.0100 Y with [Cr(II)], [Complex] and ionic strength

constant. The order of the reaction with respect to [Cr(II)] was

confirmed by use of the stopped-flow instrument with acid, complex

and ionic strength constant and [Cr(II)] = 9.00 x 102 M. Results

are summarized in Table VI. In the specified ranges the reaction

was found to be first order in oxidant and reductant and independent

of the acidity. The second-order rate constant was determined to

be (2.22 t 0.12) M-1 sec"1 (Table VIII).


Activation Parameters for the Reaction Between the
Cobalt(III) Complexes and Chromium(II)


Activation parameters for the reaction of the cobalt(III)

complexes and chromium(II) were obtained by determination of the

second-order rate constants for the individual reactions at temper-

atures varying from 13C to 450C, depending on the system. A plot

of log (kznd order/T) vs 1/T was made as previously described. At
least throe determinations at each of three different temperatures

were made. Temperature was maintained at t 0.100C by methods pre-

viously described. For the reactions at other than 25.0C, solu-

tions were thermostattcd for at least one hour prior to transfer-

ence to the reaction vcsols and further therrmostatted there for








at least one-half hour prior to reaction. Rate constants so

obtained for the four cobalt(III) complexes are summarized in

Table VIII.

Average values at each temperature were used to determine

the respective activation parameters via a least squares analy-

sis. Plots so obtained are shown in Figures 2, 3 and 4. Error

limits for the activation parameters were determined from the

most deviant lines still encompassed by all rate error limits.

Results are summarized in Table IX.

Attempts to determine activation parameters for reaction

involving the deprotonated form of the Co(III) lactate complex

were abandoned due to a >20% fluctuation in reproduciblity at

[H+] = 8.0 x 10-5 M where the deprotonated form is dominant.

This resulted in overlapping values of kobs at the various

temperatures. At intermediate pH levels distinct curvature in

the Eyring plots was observed. This curvature could arise from

different activation enthalpies for the two contributing paths

as well as from an expected variation in K -with temperatures.

Thus, values at the lower pH values would not realistically

measure the activation energy for the deprotonated form without

a determination of Ka and the acid dependence at each temperature

which was not done.


Substitution Reactions of the Chromium(III) Products




Using a slight excess of [Co(en)2(0OCCH(Ca3)S)1+ relative to












r


2.8



2.7 -


2.51-


3.10 3.15 3.20 3.25

103/T, (K')


330 335 3.40


Figure 2. Eyring plot for [Co(en)2(ooCCH(CH3)S)]+.
M = 0.100 M (HC104--LiC101).


L_71


I I I I I




































3.15 3.20 3.25 3.30

103/T, (K-')


335 3.40


Figure 3. Eyring plot for [Co(en)2(OOCCH2SC3)]2+.
A = 1.00 M (HC104--LiCO1).


0.2-



0.1



0.0



-0.1



-0.2 -



-o0.3 1 I I I












-1.3


-1.5 -
--




-1.7-







-2.1



-2.5


-2.I I


3.40 3.45


Figure 4. Eyring plots for some cobalt(III) complexes.
-- [C.. ,-,i C -)2+
---- [Co(en)2(OOCCH(CH")OH)]2
A = 1.00 (IHC104--LiC104).


325 3.30 335
103/T, (K1')










TABLE IX
Activation Parameters for the Reactions of
Chromium(II) with Cobalt(III) Complexes

Complex HAHt S AG$ a
Kcal mole-1 cal deg-1 Kcal mole-1

[Co(en)2(OOCCH(CH3)S)] b 1.1 1.7 -31.1 5.7 1.04 t 3.4
[Co(en)2(ooCH(CH3)oH)]21+ 4.7 0.4 -39.1 t 1.0 16.4 t 0.8
[Co(en)2(OOCCH2SCH3)]2' c 8.5 t 0.8 -18.9 t 2.5 11.2 3.5
[Co(en)2(OOCCH2NH2)]2+ 8.8 08 -27.4 t 2.6 17.0 1.6

aT = 25.0oC. b = 0.10 M (HClO4--LiCIO). cP = 1.00 M (HC104-LiClo).







chromium(II) at high acid concentration, the stopped-flow
instrument can be used to measure the rate of the first reaction

subsequent to the initial redox reaction. This secondary reaction

was followed by monitoring the increase in absorbance at 545 nm,
a maximum for the chelated 2-mercaptopropionate chromium(III) com-

plex. This subsequent reaction has been postulated as the ring

closure reaction of [Cr(H20)5(S(CH3)CHCOOH)1+ (mercaptide-boud),
the initial product of the redox reaction, vide supra.

The first experimentally isolable product of the redox
reaction under all conditions has been characterized as the chelated

chromium(IIl) product of this ring closure, [Cr(H20)4(OOCCa(CH3)-

S) +, vide supra. This product underwent a subsequent reaction over
a twenty-four-hour period which was then separated via ion exchange

to yield mostly +1 and +2 ions with some small amount of a +3 ion
also present, especially a low acid concentration ( (0.20 M). The

+1 ion exhibited identical spectral parameters to the original
chelated complex, [Cr(H20)(00OCCH(CH3)S)]1, while the +2 fraction

was spectrally identical to the monodentate complex, [Cr(H20)5(0O-

CCH(CH )SH)]2. The +3 ion was spectrally identified as [Cr(H20)-

63+. By spectrophotometric monitoring of the 264 nm absorption it
can be shown that the +2 fraction converts back to the chelate form.
This peak has been assigned to a metal-coordinated mercaptide chro-

mophore on the basis of its presence in the chelate complex, ab-
sence in the monodentate carboxylate-bound complex and similarity
in energy and molar absorptivity to other chromium(III)-mercaptide
species (Table III). These observations can be understood by con-
sideration of the following net ionic reactions and observed rate







laws at constant acidity.

[Co(en)2(0OCCH(CH3)S)] + Cr + 5H -

[Cr(H20)5(SCH(CH3)cOOH)+ Co2 + Co 2enH2+ (A)
I

I k* [Cr(H20)4(OOCH(CH3)S)J] (B)

II


II + H+ [Cr(H2O)5(0OCCH(CH3SH) 2+, K (C)
III

d[II]
dt = -k II + kIII], kobs kf +

The rate of the initial process (B) to form II varied
linearly with acidity: A plot of k vs [H+] (Figure 5) where [H+]
is varied from 0.900 M_ to 0.0900 M results in the expression k =
(1.0 + 2.6 [H1]) x 102 (M and sec, 1 = 1.00 M (HCI0--LiCID4),
250c).
A determination of the Ke for reaction (C) was necessary in
order to evaluate kf and kr, vide sura. Ke was determined by
allowing a solution of II to equilibrate at known acidities. The
equilibration was followed spectrophotometrically at 545 nm, a
maximum for II, over a twenty-to thirty-hour period with readings
taken every two thousand seconds for the first twenty hours. Since
the hydrolysis of III to [Cr(I20)6]3+ marginally overlaps the
equilibration, a plot of Aob vs time was then used to evaluate the
absorbance at equilibrium. A was closely approximated by extra-
polation of the very gradually decreasing terminal portion of this










































Figure 5. Acid dependence of ring closure of
[Cr(H2o) (SCH(CH 3)COH)]2.







plot to t = 0 sec. This allowed a close approximation of Ke from
the known molar absorptivities of II and III and the acidity. The

average value for three determinations gave Ke = 10.5 (Table X).
The slow hydrolysis of III to [Cr(H20)6]3+ (k = 10-7 sec1) gives

rise to a small inherent error in the Keq evaluations.

The evaluation of kf and kr was accomplished starting with
solutions of II for the sake of convenience. The observed conver-

sion of III to II could have been used alternatively. The acidity

range examined was dictated by side reactions at higher pH to be
examined further and discussed elsewhere.

The evaluation of kobs at a given acidity was accomplished by
plotting log (At- Aeq) vs t. The plots were found to be linear

over at least two half-lives. For reactions with [H+] < 0.30 M,
Ae was calculated from the value of K the acidity, initial con-

centration of II and the respective molar absorptivities of II and
III. Values of kf and kr were then calculated from kobs, Ke and

CH] by methods previously described.
Within the range [H+] = 0.65 M to 0.100 M, kf was found to
vary linearly with acid concentration while kr was independent of
acid concentration (Table XI). Plots of kf vs [H+] (Figure 6) and

kr vs [H+]-1 (Figure 7) gave the following rate expressions:

kf = (7.31 [H+]) x 10-5 (M and sec)

kr = 7.10 x 10-6 (M and sec)

Error limits for the respective plots of kf and kr were obtained

from the error limits in Keq and evaluation of kf and kr based on
the maximum deviation values.












TABLE X
Evaluation of Equilibrium Constant for the
Chromium(II )-2-Mercaptopropionate Interconversiona


[Hf] [CrL] [CrL2+l] K
M M x 103 Mx 103


1.00 0.11 1.29 11.7
0.500 0.23 1.17 10.2
0.300 0.35 1.05 9.7
Avg = 10.5 1.2

a/A = 1.00 M (HClO4--LiC04), 25C, 100 ml solution.




70


TABLE XI
Acid Dependence for Interconversion of the
Chromium(III )-2-Mercaptopropionatea


[H+ kb k kr
M x 105 sec-1 x 105 sec-1 x 106 sea-1


1.00 9.34 8.51 t .05 8.11 t .67
0.650 5.85 5.12 t .05 7.45 t .60
0.500 4.26 3.57 .05 6.83 .52
0.300 2.95 2.24 .05 7.12 .50
0.180 2.04 1.34 t .04 7.07 t .44
0.100 1.47 0.75 .04 7.16 .37

a, = 1.00 M (HC104--LiCO04), 250C.






























[H ]


Figure 6. Acid dependence of ring opening of
[cr(oT20) ( -" ". ''.- ,)l .

































CI I I
0 2.5 5.0 7.5
[H]


Figure 7. Acid independence of ring closure of
[Cr(H2o)5(OOCCHI(CH,3)HH)]2+.








When excess chromium(II) was used in the initial reaction of

chromium(II) with [Co(en)2(OOCCH(CH3)S)] no subsequent spectral

change indicative of the conversion of I to II was observable on

the stopped-flow instrument. Duplication of the conditions for

product studies yielded the chelated 2-mercaptopropionato-chromium-

(III) complex in virtually stoichiometric amounts. These observa-

tions indicate that there is, in the presence of excess chromium(IIU

another reaction of the mercaptide-bound pendant intermediate.

This is most likely a second oxidation-reduction reaction in which

chromium(II) reacts with the monodentate chromium(III) product

initially formed to produce chelated chromium(III) product and re-

generate chromium(II). In view of the latter product, this can be

considered as a chromium(Il)-catalyzed chelate ring closure. Simi-

lar observation have been made for the analogous 2-mercaptoacetato

chromium(II) and maleato chromium(III) systems, respectively.10'30

Lactate as Lifand

The lactate ligand system was included in this study primary

for comparative purposes in the redox relations with attention fo-

cused primarily on the sulfur analogue. As such, detail of investi-

gation into the chromium(III)-lactate system is less than that

previously described for the 2-mercaptopropionate system, but the

experimental results obtained suggest that future exploration of

the oxygen system merits consideration. The following observations

delineate the limits of our investigation.

As previously discussed, the monodentate alkoxide-bound

chromium(III) species, [Cr(H20)5L ('L.ln (i )Lu) J has not been







experimentally observable as a product of the Cr(II)-[Co(en)2(00-
CH(CH 3)O)] reaction at high pH and of the path inverse in acid at
low pH. The observable products, identified by chromium analysis,
ion exchange chromotography and spectral studies as [Cr(H20)5(OOCC-
H(CH3)OH)]2+ and [Cr(H20)4(OOCCH(CH3)0)]+, arise in varying ratios
which are dependent on the acidity conditions. Further, the iso-
lated species are interconvertible as a function of the acidity.
Only the blue, +2 ion is produced under conditions compar-
able to the following: [H+] = 0.100 M, [Co(III)] = 0.10 0.010 N
with chronium(II) ion in excess, stoichiometric or deficient amounts.
The ion-exchanged +2 ion, maintained throughout at pH = 1, yielded
the following spectral parameters: [A(e): 568(26.8), 413(33.2)].
Dropwise addition of 1.0 M NaOH to this solution to 2.8 = pH = 3.6
(via pH meter) produced a pink-orange species whose visible spec-
trum had the following characteristics: [A(): 548(31 1.5,
438(38 t 1.6)]. Reacidification of the solution to pH = 1 regen-
erated the blue species originally obtained, vi., [xA(): 568(25
1), 413(34 1)]. If, instead of using an eluent of pH = 1, a
neutral eluent is used, the blue ion elutes as a +2 ion but turns
pink immediately on coming off the column. The pH of this solution
was found to be 2.8 and the visible spectrum was identical to that
previously characterized as the [Cr(H20)4(OOCCH(CIH )o) ion.
Acidification of this solution to pH = 1 regenerates the previously
described [Cr(H20)5(0OCCII(CH3)01)]2+ species as identified by its
visible spectrum.
In order to test the possibility that the conversions were
sioply, due to proton transfer, which should be extremely rapid,








stopped-flow kinetic runs were carried out. A solution of the

+1 ion, generated by conversion of the +2 ion through appropriate

adjustment of the pH to 3.0,was reacted separately with equal

volumes of 0.100 M and 0.200 M HCIO4 (p = 0.25 M_ (HC104--LiCI04)).

Concentrations after mixing were [Cr(III)] = 1.0 x 10-3 M. [H+ =

0.050 M and 0.100 M, respectively. Spectrophotometric monitoring

was at 438 nm, a maximum for the chelate species. A plot of

log(At- A,) vs t gave a first-order rate constant of kobs = 3.2 x

10-2 sec- for both acidities, showing the overall rate process to

be measurable. Thus, simple proton transfer appears excluded.

The lower limit of acidity used for the redox reaction (4.0
-4
x 10 M) was dictated by the release of two moles of ethylenedia-

mine per mole of oxidant. Upon reaction the pH increases due to

the consumption of protons by the amine functions thereby intro-

ducing the hazard of metal hydroxide precipitation. Using this

initial acidity with stoichiometric amounts of chromium(II) ion

and cobalt(III) complex results, upon cation exchange separation

of the products using neutral eluent, in isolation of a +1 and a

+2 ion in the molar ratio 2:1. The ions were characterized spec-

trally as the [Cr(H20)4(OCC(CH33 )0)]+ and [Cr(H20)5(OOCCH(CH3)-

OH)]2+ species. Again variation in acidity produced interconver-
sion of the ions as previously observed.

Methylthioacetate as Ligan

The relationship between this ligand and the mercaptoacetate
ligand, of which it is a derivative, and the similarity of their

cobalt(III) complexes invites a comparison of the behavior of the

chromium(III) products. Relative to the previous results with








mercaptoacetate, any differences would be directly ascribable to

the transformation of the mercaptide function to a thioether func-

tion.

The only isolable product of the initial oxidation-reduction

reaction in the acidity range 0.100 M 4 [H+] 1 0.010 M and with

chromium(II) in excess, stoichiometric or deficient amounts was the

carboxylate-bound chromium(III) product [Cr(H20) 5(OOCCH2SCH ,) ,

the characterization of which has previously been described. The

ion underwent no reaction of interest other than hydrolysis to the
[Cr(H2O)6]3 species; therefore no further work with this ion was

undertaken.

By employing techniques analogous to those used for the thio-

lactate complex, the author hoped to be able to discern formation

of a thioether-bound intermediate, thus confirming the bridging

ligand as the thioether rather than carbonyl oxygen. Reactions

were performed using the all-glass mixing apparatus and the Cary 14

instrument. The reactions were monitored at 530 nm, an absorbance

maximum for the chromium(III) chelate complex (by analogy with sim-

ilar complexes, Table III), and at 270 nm, the spectral region of

greatest difference in molar absorptivities for the sulfur-bound

species relative to the carboxylate-bound monodentate chromium(III)

product. The respective experimental conditions were as follows:

530 nm; [co(III)] = 5.10 x 104 M, [Cr(II)] = 4.7 x 104 M_, [H+] =

0.100 M and = 1,00 E (i 'i.. LICi.: ), 270 rn; [Co(III)] = 2.58 x

10 M, [Cr(II)] = 2.35 x 10 4, [H+] = 0.100 M and I = 1.00 M

(HC104--LiClOJ4). In both cases a rapid decrease in absorbance
corresponding to 90 % reaction for the oxidation-reduction








was followed by a slow, small decrease in absorbance. The plot of

log (At- A,) vs t was characteristic of two consecutive first-order
reactions subsequent to the initial oxidation-reduction. A plot of

(b l)Ao ()A, + A
log a a- vS time
At A,

for the initial rapid decrease in absorbance was characteristic of

a second-order reaction corresponding to reaction of [Co(en)2(OOCC-

H2SCH3)]2+ with Cr(II).
The subsequent consecutive first-order reactions can be under-

stood in terms of a two-step mechanism consisting of: (1) closure

of a first-formed sulfur-bound monodentate chromium(III) product

to yield the chelate [Cr(H20)4(OOCCH2SCH )]2 followed by; (2)

opening of the chelate ring to yield the carboxylate-bound monoden-

tate [Cr(H20)5(OOCCI2SCH3)]2+ species. The subsequent first-order

reactions observed can be rationalized only with great difficulty
if redox bridging is postulated to proceed via carbonyl oxygen.

Glycinate as Ligand

The inclusion of this ligand system in the present study was,
as in the lactate case, primarily for purposes of comparison in

interpreting reactivity patterns. As such, detailed investigative

work into the products of the reaction of chromium(II) with [Co(en)-

2(OOCC2H22)2+ was not carried to the extent of the 2-mercapto-
propionate system. As will be discussed below, limitations imposed
by the system itself hinder complete work, but certain salient
features of the chromium(III) glycinate product were accessible.
In the range of acidity used (0.100 j 1 [H1 ] H 0.0100 M) the

reaction of [Co(en)2(OOCCQi2i2) 2+ with chromium(II) used in excess,








stoichiometric, or deficient amounts, the only isolable product

was characterized as the ion [Cr(H20)5( oCCH NHI )] with spectral

parameters as follows: [\(e): 573(22 1.1), 411(23.0 1.6)].

The ion was eluted from a cation exchange column in the lithium

form with 1.0 4 LiC104. The pH of the eluted solution was found

to be 3,5 + 0.1. Upon addition of 1.0 M Na0H solution to the

product solution until pH = 4.5 t 0.1, the solution color changed

immediately from blue to green, but yielded virtually the same

visible spectrum as the original solution. After several days
the color had chanCed to a red-violet and separation was effected

using ion exchange techniques. A fraction presumed to contain a

+2 ion was collected which exhibited the following visible spectrum:

[A(E): 55(38.0 5), 420(41 + 6)] (pH = 4.5). The large error
limits for the molar extinction coefficients are a result of the

dilute solutions (10 M) necessarily employed. From the rather

high values for the coefficients in comparison with monodentate

carboxylate-bound chromium(III) species (Table II), the ion is pre-

sumed to be [Cr(H20)4(00CCH2M2)]2+. The important features are

that a 1:1 glycine-chromium(III) product can be isolated from the

appropriate p1, undergo a subsequent reaction to yield, in part
a chromium(III)-glycine chelate complex.













DISCUSSION


The primary objective of this research was to better

define the influence which coordinated sulfur functions have on

the reactivity of metal complexes in oxidation-reduction reactions.

Conclusions relating to this objective will be discussed first.


e,,hF tionr of ti,-e I:.: iII) Ci: r.ol xes L, L-.r orLUim )


All redox reactions between the cobalt(III) complexes and

chronium(II) were demonstrated to proceed by inner-sphere pathways

through product analysis.

In attempting to understand the rate and activation energy

data to be presented, it is convenient to describe the net process

for an inner-sphere electron transfer reaction as series of

steps15,4 represented by the following equations (for clarity, only

the bridging ligand is represented):

kl
Co(II)X + Cr(II) k Co(III)-X-Cr(II) K1 (14)
k_-l 1

k2
Co(III)-X-Cr() Co(III)-X- )] K2 (15)
k-2


[Co(III)-X-Cr(II)J* ; [Co(II)-X-Cr(III)]* (16)
k3









k4
[Co(II)-X-Cr(III)]*- Co(II)-X-Cr(IIl) (17)



Co(II)-X-Cr(III) Co(Il) + X-Cr(III). (18)
-5

Equation (1) represents the substitution equilibrium between the

reactants and bridged precursor complex which can rearrange to the

activated complex (2). Electron transfer is represented by equa-

tion (3) while equations (4) and (5) represent subsequent deacti-

vation of the successor complex and decay to products, respective-

ly.
The rate of formation of the precursor complex can in certain

cases be rate determining. If the collision rate for the positively

charged complexes is taken to be 109 -1 sec-1, the lifetimes of

the resulting outer-sphere encounter complexes estimated as 10 -
-12
10 sec, and the rate of exchange of a water which is coordinated

to chromium(II) and proximate to the bridging ligand is 10 1010

sec-1,7ab an estimate of 107 M-1 sec-1 is obtained for kl.7c Should

k1 not be Lhe rate determining step, the stability of the precursor
complex becomes important as an equilibrium prior to the rate-deter.

mining step, formation of the activated complex. In this case the

free energy of activation can be expressed as AGt = -RT ln(K1K2).

This enables a discussion of the reactivities in terms of steps (1)

and (2) whether or not they are actually isolated in time.

This model provides a basis for the discussion of the reacti-

vity parameters obtained in this study together with previous

results which are included for purposes of comparison (Table XII).




TABLE XII
Reactivity Parameters for the Reaction of Some Cobalt(III)
ComplexeS with Chromium(II)


Entry Species Mechanism kCr(II) AH ASb AG* Ref.
M- sec- Kcal mole eu Kcal mole-


1 [Co(NH )5 (00CH)] 2
2 [Co(NH3)5(ooCCH3)1]2
3 [Co(NH )5(oocc(cuH )]2
4 [co(NH3)5(ooCCHNH3)13+
5 [co(NH3)5(OOCCoH2H)]2+
6 [Co(NH3)5(00CCH(CH 3)H)]2+
7 [Co(NH)5 (OOCC(CH3)20H)]2+
8 [Co(en)2(OOCCI2NH )]2+
9 [Co(en)2(OOCCH2SCH 3 )]2
10 [Co(en)2 (H2CH2CH2SCH3)]3+
11 cis-[Co(en)2(OOCH)2
12 [Co(en)2(OOCCH20)]+
13 [Co(en)2 (OCH2CH2NH2)]+
14 [Co(en)2(OOCCH2S)]+ e
15 [Co(en)2(NH2CHHH2S) 2+
16 [co(NH3)5(py)]3+
17 [Co(NH3 )5(nicotinamide)] 3
(pyridine-N-bound)


i(COO) 7.2 8.3
i(coo) 0.35 8.2
i(COO) 0.0096 11.1
i(COO) 0.06 7.7
i(COO) 3.1 9.0
i(COO) 6.65 --
i(COO) 11.5 9.1
i(COO) 2.22 8.8
i(S) 274 8.5
i(S) ? 0.38 5.4
i(COO) 50 7.9
i(0) 1.9 x 103 ( 2.2)
i(0) 935 5.1
i(S) 6.4 x 106 (l.l)est
i(S) 3.5 x 104 7.3
o 0.0043 9.8
i 0.033 10.2
o 0.014 9.2


-27 16
-33 20
-31 20
-38 19
-26 17


-24 16
-27.4 17.0
-18.9 11.2
-42.4 18.0
-25 15.4
( -36) (13)
-28 13.6
(-24)est (6.9)est
-13.3 11.2
-36 21
-31 20
-36 20





TABLE XII continued

Entry Species Mechanism kCr(II) AHR AS* AG* Ref.
M- sec- Kcal mole- eu Kcal mole

18 [Co(NH ) (4-pyridone)]3 i 0.014 -- -- h
(oxygen-bound) o 0.0096 h
19 [Co(NH )5(OC(NH)2)13+ o ? 0.019 10.6 -31 20 i
20 [Co(NH )5 (NCNC(NH2 )2)1]3 o ? 0.029 8.3 -37 19 i
21 [Co(en)2(OOCCH(CH O)) i(0) 52 (2.3)est (-43)est (15)est c
22 [Co(en)2(OOCCH(CH )S)]+ e i(S) 1.55 x 105 1.1 -31.1 10.4 c
23 cis-[Co(en)2(C1)(H 20)+ e i(Cl) 3.8 x 105 1 -29 9.8 44c
24 cis-Co(en)2(F)(H20)]2+ e i(F) 1.4 x 105 0.0 -34 10.4 44c
25 [Co(NH3)5(oH)]2' i(0) 1.5 x 105 4.6 -18 10 j

ajA = 1.00 M. b25.0oC. CThis work. dJ.F. Ward and A. Haim, J. Amer. Chem. Soc., 92, 475(1970). epk =
0.10 M. R. G. Linck in "Reaction Mechanisms in Inorganic Chemistry," M.L. Tobe, Ed., Medical and Techni-
cal Publishing Co. Ltd., London, In press. F. Nordmeyer and H. Taube, J. Amer. Chem. Soc., 90, 1163(19681
hE S. Gould, J. Amer. Chem. Soc., 90, 1740(1968). iE. J. Balahina and R. B. Jordan, J. Amer. Chem. Soc.,
93, 625(1971). JA. Zwickel and H. Taube, J. Amer. Chem. Soc., 81, 1288(1959).








For clarity the results will be discussed according to the sequence

of bridging functions (1) chelated carboxylate, (2) chelated

thioether (3) chelated thiolate and alkoxide. The order of

entries in Table XII is that in which they are encountered in this

discussion.


Chelated Carboxylate as a Bridging Function


All the complexes studied contained as potential bridging

functions coordinated carboxylate groups in bidentate ligands with

the other donor function also coordinated to the same metal center.

This situation represents a departure from previous studies in

which simple monodentate carboxylate ligands were examined or in

which a potentially chelating donor function remained pendant from

the carboxylated metal. It was, therefore, deemed essential to

establish any distinctions in bridging efficiency between the

carboxylate group of a chelate and those previously studied. For

this reason the reaction of [Co(en)2(OOCCHB2N2)j with chromium(Il)

was investigated.

Entries 1-7 of Table XII summarize prior experimental results

for the types of carboxylate coordination previously studied. The

acetato complex, entry 2, can be taken as the prototype. The more

rapid rate of reduction observed for the format complex, entry 1,

resides in the entopy component and can be attributed to a dimin-

ished steric restriction15 which is expected to be concentrated in

a greater stability of the precursor complex, KI. While the rate

of the more sterically hindered iso-butyrato complex, entry 3, is








diminished as expected, the source of this ditmi tion is found

unexpectedly in the enthalpy of activation, a result not understood

by this author. The diminished rate for the pendant glycinato

complex, entry 4, is attributable to the entropy difference, a

logical consequence of the increased charge.15 Entries 5, 6 and 7

are for complexes with pendant functions which can chelate the

chromium(II) reductant. The expected greater stability of the

precursor complex and enhanced rates are reflected in more favor-

able entropy contributions.

In this context the enhanced rate for the chelated glycinato

complex, entry 8, appears easily understood. The acceleration

finds its source in the entropy term. This is ascribed to a

greater stability of the precursor complex when the carbonyl func-

tion to which chromium(II) most probably binds15 is held in a more

accessible position as a result of chelation by the amine function.

This effect is expected to extend, with allowances for variations

in charge type, to other chelated carboxylate ligands, thereby

fulfilling one objective of this research.


Chelated Thioether as a Bridging Function


For the important case of thioether coordination in [Co(en)2-

(OOCCH2SCH 3) entry 9, it was not possible to define the bridging

function through isolation of the chromium(III) product which was

always found to be [Cr(H20) I",-CC'"V'F )]. While this result

appears to indicate carboxylate bridging it is important to con-

sider the initial product of the alternative sulfur-bridged path,








[Cr(H20)5(CU3SCH2COOH)]3+. Such a consideration is mandatory in

view of a rate substantially greater than that found as typical for

carboxylate bridging even in a chelated example of the same charge
45
type, video supra. Further, previous research suggests that

chelate closure of the alternative product to yield [Cr(H20)4-

(CCH3CH2C)]2+ could occur in times shorter than those required

for isolation at the acidity level (0.100 M) of our experiments.

A relatively rapid hydrolysis of the chromium-thioether bond, which

would not be surprising, would lead to our product observations.

In this context it should be noted that the relatively rapid

redox rate observed arises exclusively from an entropy contribution

which is -12 eu more favorable than for any previously studied

carboxylato-bridged reaction lacking a pendant donor function.

Further, the absorbency changes at 530 nm and 270 ra reveal a

sequence of three steps which cannot be ascribed to carboxylate

bridging, a mechanism which should result in a single-step absor-

bency change. However, the two substitutional processes previously

described for [Cr(H20)5(CH3SCH2COOH)]3+ could account for two

absorbency changes subsequent to the redox step. These results

are taken as indicative of bridging via the thioether function.

If this interpretation is correct, the inner-sphere reactivity

bestowed by a coordinated thioether is one of very few examples

lying intermediate between that bestowed by very efficient bridg-

ing ligands, e.g. the halides and thiolates, and by the rather

mediocre bridging ligands, e.g. water and carboxylate. Thus, a

second major objective of this study seems fulfilled.








A detailed discussion of the reactions of the [Co(en)2-

(CH3SCH2CH2N H2)]3 complex, entry 10, is appropriately deferred to

another thesis. However, its greater diminished reactivity is

attributable to a substantial decrease in the entropy term while

the enthalpy actually contributes in the opposite direction. In

fact, its reactivity parameters, in comparison with those for the

complex described here, are decidedly at variance with those anti-

cipated for an inner-sphere reaction in which the non-bridging

function cis to the bridging ligand has been changed from carbox-

ylatc to arine. Such a ch'rie results in little variation in all

parameters for one comparison, entry 11 vs entry 1, and little

variation in the observed rate for another, entry 12 vs entry 13.

A somewhat different pattern emerges for two other inner-sphere

reactions entailing a similar variation of the cis non-bridging

function, entries 14 and 15. The reasons for this different

pattern are not yet understood. However, if the more rapid rate of

reduction for the [Co(en)2(OOCCH2S)]+ complex is ascribed to en-

thalpic and entropic variations similar to those used for the thio-

ether comparison a value for the entropy of activation is obtained

which seems unrealistically high. Thus, a different comparative

pattern from those previously observed for consistently inner-sphere

reactions is evident. The results reported here may prove useful

in assigning the reduction of [Co(en)2(CH SCH2CH2NH2)]+ to the

outer-sphere category. The activation parameters are in substanti-

al agreement for other outer-sphere reductions involving +3 ions,

entries 16-20, although tis, in itself, is not diagnostic of the

mechanism.








Thiolate and Alkoxide as Bridging Functions


In the pioneering research in this area, Lane found a

reactivity for coordinated thiolate as a bridging function which

exceeded that for a comparably coordinated alkoxide by a factor

of >3,000.10 Three possible reasons for this enhanced reactivity

were presented: (1) a greater stability of the precursor complex

with the thiolate ligand arising from the greater steric accessi-

bility of the sulfur atom, (2) a cobalt-sulfur bond which is

weaker than the cobalt-oxygen bond,thereby requiring less enthalpy

for activation of the precursor complex, (3) a possibly greater

sigma covalency in the cobalt-sulfur bond which might contribute

to an enhanced probability for electron transfer. No distinction

was possible between the relative contributions to the reactivity

from these sources.

The initial phases of this study were directed toward

providing such a distinction. Space-filling models suggested that

the methylene hydrogens on the carbon atom adjacent to the coordin-

ated chalcogenide would inhibit precursor complex formation with

[Cr(H20)5?+ to a greater extent for alkoxide than for the larger

thiolate sulfur. According to this view, it was felt that substi-

tution of one or two methyl groups on this carbon atom would dimin-

ish the rate of reduction for both complexes via a steric effect

without drastically altering the electronic contributions. (Unfor-

tunately attempts to prepare the dimethyl derivative were unsuccess-

ful,but the monomethyl derivative proved to be synthetically access-

ible.) It was naively assumed that the reactivity of the thiolate








complex would be less sensitive to this change than the complex

with a smaller, less accessible oxygen-bridging atom. Finally,

if the expected diminmtions in rates actually materialized it was

hoped that a determination and comparison of the activation para-

meters would be possible.

The anticipated decrease in the rate of reduction was, in

fact, observed, as a comparison of entries 12 and 21, 14 and 22

indicates. Surprisingly, the factor by which the rate is decreased

is -1/40 for both the alkoxide and thiolate complexes. This im-

plies that any greater anticipated steric susceptibility to inhibi-

tion for the alkoxide function compared to that of the thiolate is

not developed by monomethyl substitution. For both complexes the

decrease in the accessibility of the bridging atom would, in the

absence of sufficient activation parameter data, appear to be

comparable implying a highly directional approach for the [Cr(H20)-

5+ residue (i.e., the methyl function exerts a restrictive influ-
ence but can be comparably avoided in both case).

In the case of the mercaptopropionate complex it was possible,

as the anticipated result of the decrease in rate, to measure the

activation parameters. The enthalpy of activation of 1.1 kcal/mole

reflects an unusually small resistance to reaction from this factor.

The reason for the difference in this parameter in the reduction of

the mercaptoethylamine example, the only other case for which it

has been determined, is not presently understood. The latter com-

plex nay be anomalous with regard to the enthalpy contribution.

Its anomalous absorption spectrum in the visible region suggests
something unusual in its electronic configuration. Even if this
something unusual in its electronic configuration. Even if this








were not the case certain differences should be recognized in the
2
two ligands. The mercaptopropionate ligand possesses an sp carbon

in the chelate skeleton whereas the mercptoethylamine has only

sp carbon atoms. This difference could result in different con-

formations for the two chelated ligands. The consequences for the

activation parameters of such variations in chelated ligands is

essentially unexplored. Further speculation is best postponed un-

til more data are available.

The observed entropy of activation for the mercaptopropionate

complex seems remarkably positive for a species with a methyl group

and a hydrogen atom on a carbon atom bound to the bridging atom and

further constricted by the chelation of the ligand. In fact, the

steric restrictions appear to leave the sulfur as accessible as the

halide-bridging ligand in cis-Co(en)2(Cl)(H20)]2+ and s-[Co(en)-

(F)(H20)]2+ and, in spite of the methyl substitution, less restrict-

ed than the alkoxide oxygen of the unsubstituted glycollate complex,

vide infra. These entropy trends are taken as indicative of a

uniqueness of the large coordinated sulfur atom in remaining

sterically accessible in spite of rather bulky substitutions.

The activation parameters reported in Table XII for the gly-

collate complex, entry 12, are to be regarded as tentative and

subject to confirmation. Nevertheless, they appear reasonable in

that AH lies intermediate between that for a mercaptide bridge

in a comparable environment, entry 22 (compare also entries 23

and 24), and that reported for [Co(NH3 )5OH)J which should have

a higher AH* as the result of the change in non-bridging ligands;

compare entries 1 and 11. The value for AS seems appropriate.








The discrepancy between the tentative value for AH of 2.2 kcal/

mole and the 5.1 kcal/mole observed for the reduction of [Co(en)2-

(OH2CH2CUH12)]+ may arise from an enthalpic anomaly for the latter

similar to that suspected for the [Co(en)2(SCH2CH2NHel2) analogue.

If these are viewed as anomalous activation enthalpies it seems

possible to obtain tentative estimates for four unknown activation

parameters which appear reasonable, internally consistent and in

satisfactory comparison in the entropy term with the measured val-

ues for the anomalous complexes.
The estiration proceeds as follows. Since the effect of

methyl substitution on the adjacent carbon is primary steric and is

comparable for both alkoxide and thiolate complexes, it seems rea-

sonable to attribute the rate of decrease to the entropy term. If

the value of 1.1 kcal/mole for AH* in the mercaptopropionate reac-

tion is used for the mercaptoacetate reaction a value for AS* of

-25 eu is obtained. The increase from -31 eu seems reasonable for

the loss of the methyl substituent. Further, the increase from

-36 eu for the oxygen analogue is comparable to the 14.7 eu in-

crease observed for a similar change in going from [Co(en)2(OCH2-

CH2NH2) 2+ to [Co(cn)2(SCH2CH2NH2) 2. Proceeding in the reverse
direction, 6H* of 2.2 kcal/mole for the glycollate complex is

assumed for the lactate complex. This yields a value of -43 eu

for AS* which is again .12 eu more negative than for the sulfur

analouic and -7 en rore negative than for reaction with the complex
not substituted by a methyl group.

To the extent that these estimates are reasonable the follow-

ing tentLtive conclusions for the inner-sphere reaction can be








drawn. (1) The carboxylate-chalcogenide chelates with an sp2

carbon atom in the five-membered chelate ring bestow an entropic

barrier to activation which is about 8-10 eu more negative than

for the amine-chalcogenide chelates with no sp carbon. This

difference seems reasonable in view of possible conformational

differences mentioned earlier. (2) The substitution of a methyl

function on the carbon adjacent to the chalcogenide atom in the

carboxylate-chalcogenide chelate increases the entropic barrier

to activation by 6-7 eu. (3) The substitution of sulfur for oxy-

gen in otherwise analogous alkoxide complexes lowers the entropic

barrier by 12-15 eu while the enthalpy decrease contributes about

one order of magnitude (1.4 kcal/mole) to the reactivity. Thus,

at least for the carboxylate-chalcogenide ligands, the enhanced

reactivity on substituting sulfur for oxygen seems to derive about

35% from a lowering of AH* and about 65% from a more positive AS*.

Therefore, the steric component associated with the larger sulfur

atom appears to be larger than the electronic contribution to the

observed enhancements. Thus, within the limitations expressed

earlier, a third objective of the research seems reasonably ful-

filled. In this regard we wish to acknowledge our indebtedness

to the research of Robert H. Lane and Michael J. Gilroy without

which the necessary comparisons would not have been available.



Substitution Reactions at Chromium(III)




As previously described, the substitution behavior of the