Electron transfer reactions between chloroaurate(III) complexes and various inorganic and organic reducing agents in acid medium - a brief review

: The review narrates the electron transfer reactions of some chloroaurate(lll) complexes in acid medium re ported since 1970 onwards. The reactions involving inorganic ·substrates and gold(m) were studied mostly in HCI me dium whereas the oxidations of different organic substrates were investigated in sodium acetate-acetic acid buffer me dium. Increase in [H+I decreases the rate of reactions. At higher concentrations of [H+] (ca. 2: 0.5 mol dm- 3), AuCJ 4-or IIAuCJ 4 is the reactive gold(m) species whereas when the reactions were studied in much lower acidities of [H+] ( << 0.5 mol dm- 3) or in sodium acetate-acetic acid buffer medium, AuCI 4 -, AuCI 3 (0H 2) and AuCJ 3 (0H)- are the re active species which oxidize the substrates. Kinetic evidence for the participation of AuCI 4-, AuCI3(0H2) and AuCJ3(0H) is provided. The reactivity of these different gold(m) species follow the order : AuCJ3(0H)- > AuCI 3 (0H 2) > AuCJ 4-. Influence of ionic strength was noticed in the oxidations of some inorganic and organic compounds by gold(m). The rates in the presence of different salts like NaCI, NaCI0 4 and Na 2 S0 4 at constant ionic strength was found to be same in the oxidations of a number of inorganic compounds indicating that anions do not have any inlluence on the rate. The rates were found to depend not only on the ionic strength hut also on the concentration and nature of the cations. The pseudo-first-order rate constant increases in the order u+ < Na+ < K~ < Rb+ < cs+ in the oxidation of a number of inorganic substrates. The reactions exhibit diverse mechanistic behavior. Kinetic and mechanistic features associated with such reactions have been discussed and analyzed. An attempt has been made to correlate mainly our experimental observations.


Introduction
Gold is widely distributed in nature and the chemistry of gold remains an active research area 1 <a>. Gold is not essential for any Ii\·ing organism though some plants concentrate the element 1 <b>. Gold is known to exist in different oxidation states such as + 1, +2, +3, ( +4) and ( +5) for which the electronic configurations are 4 10 , cfJ, /', (d 7 ) and (~). respectively. Of these different states, the underlined states are the most important and the states in the parenthesis are unconfirmed. Complexes of different oxidation states of gold have been discussed and incorporated by Puddephatt 2 and other workers 3 . Gold(III) forms a number of stable complex ions. The anionic complexes 4 like AuF4, AuCI4, Au(CN)4 and Au(N0 3 )4 and four coordinated cationic complexes 5 such as [AuCI 2 py 2 ]Cl, [Au(phenhC1 2 ]CI and [Au(bipyhCI 2 ]CI have been reported in the literature.

762
The lowest and highest coordination number found in coordination compounds of gold are two and six respectively with the intermediate number four being the most frequent for gold(m) and two for gold(I) compounds. A wider range of coordination numbers (two to twelve) exists in organometallic cluster compounds.
Some gold(I) compounds are biologically active and used as anti inflammatory drugs in the treatment of rheumatoid arthritis 6 -9. Several gold(III) compounds were also used earlier in the treatment of tuberculosis 10 whereas a few gold(III) complexes are found to have antitumour activity 11 • 12 • Again such gold compounds are found to have severe toxic effects 13 • 14 on human health such as kidney damage and blood disorders. Such toxic effects are presumed to be related to gold(I)-gold(III) oxidation in vivo followed by gold(III)-biomolecule interactions IS( a).
Redox and ligand exchange reactions of potential gold(I) and gold(III)-cyanide metabolites under biomimetic con-ditions have been reported 1 5 (b). There are reports 15 <c> on the investigation of a metal-centered porphyrine in nonpolar solvent where a long-lived charge separated state of the product has been identified. In view of all the above findings, kinetic studies on the reduction of Au 111 by various inorganic as well as organic compounds have become increasingly interesting. This has become important from another point of view that gold(III) can behave both as a one-electron and a two-electron transfer oxidant.
Considerable amount of work has been done 16 -3 0 on the oxidations of different reducing substrates by chloroaurate(III) complexes under different experimental conditions but no attempt has been made till date to record the progress in this field. This necessitates a review of redox reactions involving chloroaurate(III) complexes and different inorganic and organic substrates and an attempt has been made to analyze and correlate the results.

Results
The standard reduction potentials relating to tlifferent oxidation states of gold at 298.1 K have been reported31, 32. The values commonly used are mentioned below.

Reaction
Eo, ( The standard reduction potentials are highly dependent on the ligands attached and those where ligands (L) like H 2 0, CI-, Br-, 1-, CW and SCW are attached to the metal with different oxidation states have been reported33.

Different oxidizing species of gold (III) :
Gold(III) complexes are stable mainly in acid medium. However, gold(III) oxidations of different substrates which have been studied in acidic medium are kinetically dissimilar and characterized by variety of reaction pathways. A number of studies reveal that the square planar AuC1 4ion undergoes fast hydrolysis 4 (b), 34 · 35 and produces two other Auiii species, namely, AuC1 3 (Hp) and AuCI 3 (0Ht. Thus in solution of chloroauric acid. four gold(III) species, viz. HAuC1 4 , AuC1 4 -. AuCI 3 (H 2 0) and AuC1 3 (0Ht coexist and render the system complicated. Theses spe-cies are involved in the following equilibria. where K 1 , K 2 and K 3 are 1.0, 9.5 x w-6 • and 0.25.
respectively at 298 K 26 · 34 · 35 . The relative concentration of each of the above species will be controlled by the pH and the concentration of CIion in the solution.

Oxidation of inorganic compounds by gold (Ill) :
The kinetics of oxidation of a few compounds of metals such as iron(n) 16  The reduction of gold(m) by PtCil-has also been studied by Moodley and Nicoi 16 spectrophotometrically and the reaction involves a complementary two electron transfer step.

(II)
No effect of CI-ion on the rate was observed, however, the increase of both the ionic strength and pH enhance the reaction rate.
Mechanism for the oxidation of Ptll (CN)/-by gold(III) has been studied 36 \<~l by the use of stopped-flow and conventional spectrophotometric method, where the different gold(III) species viz. AuCl 4 -, AuC1 3 (0H 2 ) and AuC1 3 (0f-W were involved depending on the experimental conditions. Redox reactions are proposed to take place via chloride-bridged inner-sphere mechanisms. The steps of the reaction are as follows. However, at LCn ~ 0.25 mol dm-3 and pH -5, the oxidation occurs by AuCI 4 -and AuCI 3 (0H)-via parallel paths. In such cases, the Classistance of the redox reaction has been accounted for by the formation of intermediates of the type CI. .. Pt(CN) 4 -Cl-AuCI 2 ... x 4 -, X = CI-, OW. In contrast to this study, the investigation made by Peloso 36 (b) showed only AuCI 4 -species to be involved in the oxidation of platinum(ll).
The complementary reaction between Tl 1 and AuCI 4has been investigated by Gidd et a/.3 7 in 3.0 mol dm-3 HCI medium. Under the experimental conditions, the active species of oxidant and reductant were found to be HAuC1 4 and TIC1 2 -respectively.
Kinetics of the oxidation of different compounds of nitrogen 17 -19 • phosphorous 20 , arsenic 21 , antimony 22 , and sulphur2 4 by gold(III) have been studied spectrophotometrically in hydrochloric acid medium. The reaction of hydrazinium ion N 2 H 5 -r with Auiii 17 is first order in both [Au 111 ] and [N 2 H 5 +j. Hydrogen ion inhibits the rate of oxidation. The following steps are proposed. 764 K, HAuCI 4 ~=='=-H-r + AuC1 4 - (7) k4 N 2 H 5 + + AuC1 4 -~ products (8) The products N 2 and NH 4 + are formed by one electron transfer oxidation process whereby gold(lll) is initially reduced to gold(ll). This is supported by the decrease in reaction rate ( -72%) in 5.6% (w/v) acrylamide.
The rate expression is The value of K 1 was found to be 1.01. However, the same K 1 from distribution studies 35 of HAuC1 4 between water and diethyl ether was 1.35 and from another kinetic studies 16 by Moodley and Nicol in hydrochloric acid medium was 1.1.
The oxidation of hydroxylammonium ion by gold(III) was studied in HCI 18 (a) as well as HCl0 4 1 8(h) media. The stoichiometry of the reaction 18 (a) in HCI medium is The reaction takes place through the intermediate formation of free radical followed by its decomposition to give N 2 and is in keeping with the observation 1 8(c) made during the oxidation of this substrate by Nirv.
On the other hand when the reaction was studied 18 Cb> in HCI0 4 medium, the reaction has been shown to occur as follows.
The stoichiometry of the reaction in perchloric acid medium18 (b) was studied in which gold(m) was in excess to that of the substrate whereas the same was determined in excess hydroxylammonium concentration in HCI medium 18 <a>. Consequently even if HNO is formed (via two electron transfer oxidation mechanism) in the ratedetermining step in HCI0 4 medium. in the presence of excess substrate the possibility of formation of N 2 cannot be ruled out. HNO + NH 3 0H+-----) N 2 + 2H 2 0 + H+ (14) However, in both cases the reaction was found to be inhibited by H+ ion.
The oxidation of HN0 2 by [AuC1 4 -] 1 9 in hydrochloric acid medium was first order with respect to [Auiii] and [HN0 2 ]. H+ and Cl-ions inhibit the rate. The alkali metal ions have specific effects on the rate at constant ionic strength. The rate increases with increase in size of the alkali metal ion. The reaction appears to involve different gold(III) species, viz. AuC1 4 -, AuC1 3 (0H 2 ) and AuC1 3 (0Hf.
These three species oxidize HN0 2 in the following way, The reaction possibly occurs by a complementary mechanism via Scheme 2, although the other possibility cannot be totally ruled out simply on the basis of absence of polymerization of acrylamide in presence of the reaction mixture.

J/CS-4
Copper{II!) and silver(III) are known to oxidize sodium azide 10 • 38 in alkaline medium through the intermediate formation of hexazene to give N 2 . An attempt was made to investigate the kinetics of oxidation of azide ion by chloroaurate(III) in acid medium. Gold(m) failed to oxidize azide ion. This may be due to the formation of stable compound between these two reactants as mentioned in the literature 10 . Hypophosphite ion is oxidized by gold(m) 20 in aqueous hydrochloric acid to give phosphorous acid and Au 1 . The reaction is first order in [Auiii] and [H 3 P0 2 ]. Hydrogen ion has no effect on the reaction rate in the acid range (0.15-l.O) mol dm-3 . The reaction takes place via one electron transfer process according to the following Scheme: The oxidation of arsenious acid by gold(m)2 1 in hydrochloric acid medium proceeds through the formation of a complex between arsenic(III) and gold(n~ followed by its decomposition to give the products.
{As 111 , Auiii} ~products The rate expression has been suggested as H + ions have little effect on the reaction studied in the range (0.1-l.O) mol dm-3 of HCI. A kinetic CJ-dependence would require additional chlorides in the activated complex. Since there is no cl-dependence (at constant ionic strength) it can be explained by a bridge between the oxidant and arsenic(lll). However, the pseudo-firstorder rate constant increases with an increase in salt concentration, which may be due to an increase in ionic strength or due to a variation in the nature, and concentration of the cations. It has been observed that the enhancement of the rate in presence of the cation is in the order : K+ > Na+ > u+. The electron transfer from substrate to oxidant takes places as shown below.
A number of mechanisms by which square planar system might oxidize a substrate, have been discussed in the literature 23 · 26 · 39 with regard to gold(lll) complexes. These include (i) attack by the reductant on a ligand (e.g. ct-. HO-) with resulting atom transfer, (ii) nucleophilic attack on the metal ion with electron transfer occurring in a five-coordinate intermediate and (iii) electron transfer following ligand substitution. Annibale eta/. 23 found that the rates of oxidation of alkyl sulphides to sulfoxides by AuC1 4 -correlated directly with their basicity and inversely with steric hindrance in the alkyl chains.
The oxidation of sulphurous acid by gold(III) was stud-ied24 in the temperature range 292-310 Kin (0.07-0.40) mol dm-3 HCI medium keeping constant ionic strength by the addition of NaCI. The influence of alkali metal chloride on the reaction rate was also studied and the rate constant was found to ·increase with increasing polarizability of the alkali metal ion at constant ionic strength. The reaction has been shown to occur through the intermediate formation of complex between the reactants followed by its decomposition to give free radical and gold(II). The free radical further reacts with another gold(III) to give sulphate ion and gold(II). Gold(II) reacts further with sulphurous acid and gives dithionate and gold(I).
The kinetics of the reaction between hydrogen peroxide and gold(m)2 5 in a hydrochloric acid medium (0.01-0.2 mol dm-3 ) has been studied. From the known dissociation constant and protonation constant of H 2 0 2 , it has been shown that in this concentration range, hydrogen peroxide exists mainly as H 2 0 2 , both H02-and H302 + being insignificant. Thus in such a medium gold(III) oxidizes H 2 0 2 through the intermediate formation of free radicals and the free radical (HO;) reacts with gold(III) to give 0 2 as mentioned below.
Au III + H 2 0 2 ~ Aull + (HO;) + H +  Milgrom46(a) has reported the oxidation of 2,3,7, 8,12,13,17,18-octaethyl porphyrin by gold(III) in pyridine, in which an ethyl substituent is oxidized to a chlorovinyl group. A typical 'rhodo-type' spectrum (450-650 nm) typical of a porphyrin containing one electron withdrawing 13-substituent, confirms the reaction product. The reactions of a few gold(III) complexes with serum albumin have been investigated 46 <b) by the joint use of various spectroscopic methods and separation techniques. Both weak and strong metal-protein interactions have been observed. It is suggested that adduct formation for both of these gold(III) complexes occurs through coordination at the level of surface histidines.
The reactivity of a few alkanols and a number of aryl alcohols towards gold(III) have recently been studied 27 in our laboratory in sodium acetate-acetic acid buffer medium in the pH range 3.72-4.80. The reactions of aryl alcohols were studied in tert-butyl alcohol (15% v/v). The kinetic investigations reflect the involvement of all the three gold(III) species in the reactions. The observed kinetic isotopic effect indicates that the reactions occur via C-H bond cleavage 47 , leading to the intermediate formation of free radicals and different gold(n) 48 -50 species. The formation of intermediate free radicals is supported by the observed polymeric suspension in the presence of acrylonitrile.
The reaction obeys the following rate expression : The reactivity of alkanols towards gold(III) follows the order : ethanol > methanol > 2-propanol and that of substituted arylalcohols N0 2 > H > Cl > OMe. The unsubstituted benzyl alcohol react at a faster rate than benzhydrol.
In the oxidation 28 of some neutralized a-hydroxy acids such as glycolic. lactic, a~hydroxyisobutyric, mandelic, atrolactic and benzilic acids, gold(III) behaves as a 768 two electron transfer oxidant. Spectrophotometric evidence in the initial stages of the reaction as well as the kinetic evidence indicate the formation of an intermediate cyclic complex, which decomposes in the rate determining step to give the respective reaction products, namely, formaldehyde, acetaldehyde, acetone. benzaldehyde. acetophenone and benzophenone. The observed kinetic isotopic effect suggests that the oxidation occurs via C-C bond cleavage and not by C-H bond breaking.

SchemeS
The reaction mechanism shows that both H + and clions inhibit the reaction rate, which is supported by the following rate expression. The values of K 11 (equilibrium constant for the intermediate complex formation) and k 15 (disproportionation constant for the r.d.s.) are evaluated from the substrate effect, acid effect, and chloride effect and the values are found to corroborate with one another. The enthalpy of activation (l1lJI') is found to be linearly related to the entropy of activation (!l.gl), which suggests that a similar type of mechanism is operative in all the reactions. The fairly large and negative Ml values further support the formation of the cyclic complex. The reactivity of the acids are in the order : glycolic < lactic < a-hydroxy isobutyric < benzilic < mandelic < atrolactic.
The reduction of gold(JII) by glycolaldehyde in acetic acid-sodium acetate buffer medium has been studied 29 in the pH range 3.72-4.63. Glycolaldehyde, a dimer in the crystalline state, undergoes depolymerization, which is extremely fast in water and catalyzed by acid and base. All the different forms were detected by NMR spectra in D 2 0, however, the monomeric form becomes predominant in very dilute solution ( < 0.1 mol dm-3 ). The reaction is first-order with respect to [Auiii] as well as [glycolaldehyde]. Both H+ and Cl-ions retard the rate of reaction. Under the experimental conditions, AuCI 4 -, AuC1 3 (0H 2 ), and AuCl 3 (0H)-oxidize the substrate to give glyoxal. Due to the presence of the electron withdrawing -CH 2 0H group, glycolaldehyde exists in the hydrated form. As a result the electron transfer takes place from the alcoholic oxygen to gold(III). Irrespective of whether aldehyde form or enol form of the substrate is oxidized by Auiii to give glyoxal, the one step two-electron transfer rate determining process is corroborated by the absence of polymerization in the presence of vinyl compound.
Carbohydrates have unique biological activity, which serve as store and source of energy 51 . Of the different carbohydrates, the monosaccharides are of vast biological importance, as they are known to be involved in carbohydrate metabolism 52 . They are reported to exist in acyclic and cyclic forms and possess a higher density of functional groups than any other class of organic com-pounds53. The mechanisms of the oxidation reactions in-volving monosacchandes are of immense Importance.
The reactivity of some aldoses towards gold(III) wa!. studied in HCI medium 54 · 55 , where AuCI 4 -was the only active species of gold(III). Very recently the same reactions have been studied more elaborately in sodium acetate-acetic acid buffer medium 30 . In the latter studie!>, all the three different gold(III) species are mvolved and both H+ and CJ-ions retard the reaction rate. Under the kinetic conditions, reaction appears to involve the intermediate formation of free radicals and Au 11 • Aldoses react with gold(III) in the order : triose> tetrose> pentose > hexose. The plots of log kobs versus 'n' (where n= number of asymmetnc center) is hnear (Fig 1)  in sodium acetate-acetic acid buffer medium, AuCI 4 -, AuC1 3 (0H 2 ) and AuC1 3 (0Hr are the reactive species which oxidize the substrates. Cl-retards the rate of the' reactions. This is due to the fact that it lowers the concentrations of AuCI 3 (0H 2 ) and AuC1 3 (0Hr. Of these three species AuCI 3 (0H)-is the most reactive one. Thus the reactivity of gold(III) species follow the order, AuC1 3 (0Hr > AuCI 3 (0H 2 ) > AuC1 4 -.
The greater reactivity of AuC1 3 (0H)-and AuC1 3 (0H 2 ) over AuC1 4 -may be ascribed to the easier displacement of an OW/OH 2 group than a cl-ion 2 6. Influence of ionic strength was noticeable in the oxidation of some inorganic 1 9· 21 · 24 and organic28 compounds 770 by gold(III). The rates in the presence of different salts like NaCl, NaC10 4 and Na 2 S0 4 at constant ionic strength (0.1 mol dm-3 ) was found to be same in the oxidation of inorganic compounds indicating that al'lions do not have any influence on the reaction rate. On the other hand, the rates were found to depend not only on the ionic strength but also on the concentration and nature of the cations. The rates increase as the ionic strength increases. The reactions take place between an ion and a neutral species or between two different ionic species depending upon the substrates, which were used.
It has been suggested that alkali metal ions serve as an electron bridge and that a linear relationship exists between the experimental rate constants and polarizability of the cations. An attempt was also made to correlate rate constants with crystal radii (A), hydration energy (kJ mol-1 ), polarizability (cm-3 ) and ionic potential (kJ mol-1 ) of the alkali metal ions 21 · 24 . Although the plots of kobs (at [salt] = 0.1 mol dm-3 ) versus different physical properties are not linear except for the polarizability, the pseudo-first-order rate constant increases in the order : u+ < Na+ < K+ < Rb+ < Cs+. Thus, the variation of rate constant in the presence of different cations has been ascribed to the ion association between M + and either of the anionic gold(m) species. The ion association is higher for small alkali metal ions, thereby decreasing the ionic strength and hence the rate constant.
The oxidation of inorganic and organic reductants by gold(III) may take place by either one electron or one step two-electron transfer process. It will be interesting to compare the activation parameters of different redox processes involving gold(III). The rate constants for the oxidation of Pt 11 16 , Fe 11 16, Tl' 37 and Pt(CN)/-36 by gold(m) have been studied under different experimental conditions but at constant temperature of 298 K. On the other hand, several other reactions have been studied at different temperatures and their activation parameters have been evaluated.
The oxidations of the inorganic reductants like NzHs +, NH 3 0H+, H 3 P0 2 , H 3 As0 2 , H 2 S0 3 and H20 2 have been shown to occur through the intermediate formation of free radicals and gold(II). This was based on the determination of stoichiometry in excess substrates, polymerization reaction and experimentally obtained activation parameters. The reactions are characterized by activation enthalpy in the region (27-78) kJ mol-1 except in the oxidation of H 3 P0 2 by gold(III) where the value is 128 kJ mol-1 . Thus much higher value obtained is possibly because one electron from the lone pair of the active form Table 1. The values of second-order-rate constants (k 2 = k 0 b.f[S)), eqmllbnum constant (K) for the fast step. disproporllonauon constants (k) for the slow step and the related thermodynamic parameters for the oxidallon of some morganic compounds by gold(lll)" aThe reactions were performed in HCI medium The values of~. k and K were obtamed at 303 K unless stated otherwise The errors correspond to two least-squares standard dev1a11ons (± 2cr). bAt 308 K cAt 298 K dAt 304 K of the H 3 P0 2 is transferred to gold(III) and not by P-H bond fission. The oxidation of SbC1 3 by gold(III), which has been studied2 8 in HCl medium, has been shown to occur not through the intermediate formation of Sb 1 V but by one step two electron transfer process to give Sb v. The much lower activation enthalpy of 19.2 kJ mol-1 as compared to those mentioned above corroborates the above contention. No attempt has been made to correlate As" with Mfl since oxidations of inorganic substrates were carried out under different experimental conditions and the reactions occur via different mechanisms. Kinetic results and thermodynamic parameters for different inorganic compounds are presented in Table 1.
The oxidation of alcohols and aldoses by gold(III) occurs via C-H bond cleavage whereas a.-hydroxy acids are oxidized via C-C bond cleavage. The kinetic results and activation parameters of these two series of reactions are recorded in Table 2. The activation enthalpy is found to depend on the acid dissociation constants of a.-hydroxy acids. The pKa values 56 of some a-hydroxy acids at 298 K are known, the values are 3.83, 3.86, 3.65, 3.41, and 3.04 for glycolic, lactic, a-hydroxy isobutyric acid, mandelic and benzilic acids, respectively. The plot of !!Jill a9;ainst PK. is linear 56 (Fie:. 2). Ae:ain two different lin-ear plots of As" against Mfl are obtained in these two series of reactions (Fig. 3). The results indicate that ahydroxy acid oxidations are characterized by lower activation parameters than the alcohols and aldoses. In an oxidation-reduction between an ion of a transltlon element and a compound or ion derived from non transition element, the formation of free radical intermediate from a secQnd reactant by !-equivalent reaction is likely in some cases to be more endothermic process than the corresponding 2-equivalent reaction 57 . The exothermic process is able to lower the activation enthalpy of the one step two equivalent reaction below that of the corresponding one-electron process 58 -60 . The enthalpies as well as entropies of activation for a two electron transfer process are smaller than those obtained ·1n the studies where gold(III) behaves as one electron transfer oxidant. The smaller activation enthalpies obtained in the oxidations of a-hydroxy acids lend further support to the fact that gold(III) behaves as a two-electron transfer oxidant in this series of reactions.
A few apparent gold(II) complexes are reported 61 to be mixture of gold(I)-gold(III) oxidation states. Again, the formation of unstable gold(II) as an intermediate has also been predicted by a number of workers. However, it has been possible to stabilize this unstable oxidation state in some cases 62 -67 . In one case gold(n) is stabilized by the presence of a strong Au-Au bond in certain dimeric complexes and these complexes are diamagnetic 3 . In other cases paramagnetic gold(n) complexes are reported to be stabilized by the delocalisation of the unpaired electron over the ligands. The synthesis and characterization of paramagnetic gold(II) complexes with the ligands 0aminobenzenethiol (Habt), 0-methylthioaniline (Hmta), 1,2-di(aminophenylthio)ethane (dae) and N-(2-pyridyl methyl)-2-mercaptoaniline (Hpma) have been reported. These complexes are found to be extremely stable in solid state 772 and rea~onably stable in solution. The nature and structure of the compounds were obtained from the analysis of their EPR spectra 3 . Thus convincing evidences in favour of the formation of intermediate transition state of gold(II) as well as stable gold(ll) complexes are available in the literature as mentioned above .
Gold trichloride is dimeric in both solid and gaseous state 6 8 whereas tetrachloroaurate(III) ion exists as square planar configuration. The reducing substrates enter into the coordination sphere of planar gold(m) complex and forms an unstable intermediate transition state or complex which then undergoes electron transfer to give the products. The question of whether the reactions proceed by an outer-or inner-sphere mechanism and whether the coordination sphere of the intermediate state of gold remains intact or not cannot be answered at this stage. However, the reaction rate is much slower compared to those reactions, which occur by typical outer sphere mecha-nism69·70. It is unlikely that AuCI 4 -is reduced to AuCI 4 3 -. rather planar AuCI 4 -is reduced to AuC1 2 -. Again, FeC1 2 16 • PtCll-1 6, TlC1 37 and SbC1 3 22 have been shown to be· oxidized by AuC1 4 -to give AuCI 2 -. It is to be mentioned further that oxidation of Pt(CN) 4 2 -by AuCI 4 -, AuC1 3 (H 2 0) and AuCllOHt have also been shown to occur via chloride bridged inner-sphere mechanism. Thus gold(I) which is formed is reduced to gold metal in excess reductant or by disproportionation of the gold( I) when the reaction mixture was allowed to stand for several hours. However, no detectable amount of gold metal was formed under reaction conditions used during kinetic measurements.
1 : 10-Phenanthroline as well as 2,2'-bipyridyl complexes of gold(m) 5 (b),S(c) have been prepared and isolated in their solid states. Since there is no data to investigate the reactivities of the inorganic and organic reducing substrates towards cationic complexes of gold(III), it may be interesting to investigate such reactions in future. It is known71 that collision between like charged ions are less likely whereas between two unlike ch3:rges more likely. This is because a negative ion will find itself close to a positive ion rather than to another negative ion. As a result in the reactions between ions of unlike sign there is generally an entropy increase going from reactants to activated complex whereas for ions of like sign there is generally an entropy decrease. Such studies, therefore. using cationic complexes of gold(III) may provide a clear picture in understanding the mechanisms of reactions involving cationic as well as anionic gold(III) complexes. Benzilic acid *0.158 ± 0.01 (3.97 ± 0.13) 13 ± 4 (19 ± 13) 54 ± 2 (-116 ± 7) The pH of the reactions in acetate buffer were 4.05 for alcohols, glycolaldehyde and a-hydroxy acids reactions and for aldoses pH = 3. 72.

Experimental
The absorption spectrum of AuC1 4 -obtained from different sources and under different experimental conditions are different. Gangopadhyay and Chakraborty 72 prepared tetrachloro complex by decomposing bromocomplex with aqua regia and reported that tetrachloro complex shows two intense bands in the ultraviolet region. These were interpreted as charge transfer bands, the transfer is from halogen p to gold d orbital. It may be mentioned that they have recorded the spectrum in 1 em silica cell in Beckman DU model spectrophotometer. Spectra of AuCI 4were also recorded for different [CJ-] (0.001-1.0) mol dm-3 at various acidities 26 . The spectrum for 1.0 mol dm-3 [CI-] and pH 2.0 was found to be the same as that obtained for 1.0 mol dm-3 HCI 26 . An absorption maxi-JICS-5 mum at 313 nm (E = 4.86 x 103 moi-1 dm3 cm-1) was obtained 26 with isosbestic point at 296 nm. Moreover, the absorption spectra of gold(III) solutions in the concentration range (0.30-2.37) x w-4 mol dm-3 in 0.01 mol dm-3 hydrochloric acid were studied in our laboratory in the UV region on a Shimadzu 1700 model UV-VIS spectrophotometer (Fig. 4) and the spectral pattern was found to remain unaltered with changes in the concentration of gold(III). An absorption maximum at 313 nm and a minimum of 270 nm were in good agreement with the literature value 26 . Beer's law has been found to be valid in the concentration range studied. The addition of acid (0.01-1.0 mol dm-3 ) does not seem to have any change in the absorption spectrum at constant [CJ-]. Wavelength I nm Most of the kinetic investigations were carried out spectrophotometrically although few other studies have also been made using different techniques. The reactions when studied spectrophotometrically at lower gold(III) concentrations ( -10-4 mol dm-3 ), the monitoring wavelength was 315 nm (mostly in case of inorganic substrates). On the other hand when higher concentrations of gold(III) of -w-3 mol dm-3 were used the reactions (mostly with organic substrates) were followed at 400 nm. This is because of the formation of colloidal gold in the presence of UV light, which complicated the rate measurements, and the formation of colloidal gold increased in the presence of reducing substrate upon exposure to UV light. However. gold(lii)-iron(II) reaction was studied16 amperometrically using rotating platinum electrode as the indicator electrode and the silver-silver(I) chloride as the reference electrode. A potential of 0.8 V versus the reference electrode was used. At this potential iron(ll) is mass transport controlled and the gold(III) and gold(!) are electro inactive.