536
T. S. ELLEMAN, J. W. REISHUSAND D. S. MARTIN,JR.
VOl. 80
stability of the former, in accordance with the fact that the IIIa chelate does not form until a relatively high pH is attained. The absorption spectrum of the copper chelate of I I I a is quite similar to that of the corresponding EHPG chelate. I n the ultraviolet region there is an absorption band a t 238.5 mp which increases in intensity with an increase in pH. This is apparently due to the copper phenolate linkage which is formed as the PH increases. There is also a band a t 213 mp a t very low pH which disappears a t and above the inflection point and which probably is due to the presence of the free phenolic group in acid solution. In the visible region the similarity between the copper chelates of EHPG and I I I a are quite striking and the effects observed as a result of participation of the phenolic groups in complex formation are the same for both compounds. Like EHPG, the ultraviolet spectra of a p ' ethylenediiminodi-o-cresol (11) and of 2-(o-hydroxyphenyl)-glycine (IIIa) show a band a t about 277 nip (with E of about 2000 per cresol group) which shifts a t high PH to give a phenolate band a t 292-6 mw with E of about 3000 per cresol group. In all three compounds the spectra indicate that the ionization of the phenol group is not complete until the pH is over 12. I n the case of 2-(o-methoxyphenyl)-glycine (IIIb, not shown), this variation of spectra with pH does not occur, and the 273 mp band remains a t the same intensity ( E 2000) over the whole pH range. Comparison of the ultraviolet spectra of I1 and I I I a with the spectra of their iron chelates indicates that no significant changes are introduced by combination with the metal ion. This is roughly in accord with the titration data, which indicate that a stable chelate between the Fe(II1) ion and the ligand is not formed except perhaps a t rather high PH.
L 1
2
3
4 5 m.
6
7
8
Fig. 5.-Potentiometric titration of o-hydroxyphenylglycine a t 25' and 0.1 p (KNOJ in the absence of metal ions and in the presence of various amounts of Cu(I1) and Fe(I1I) nitrates. Numbers indicate molar ratio of ligand to metal. Abscissa, m, represents moles of base added per mole of metal.
analogous to X I I I . This effect is seen in the absorption spectra by a decrease in extinction coefficient of the Fe(II1) chelate above pH 11. The significant difference in the Fe(II1) complexes of I and IIIa brought out by the titration and spectrophotometric data is the much greater
WORCESTER, MASS.
FRA~IINGIIAM, MASS.
[CONTRIBUTION NO. 587 FROM THE INSTITUrE FOR r\TOMlC RESEARCH AND DEPARTMENT O F CHEMISTRY, IOWA S r A l E COLLEGE. WORK WAS PERFORMED I N THE AMESLABORATORY OF THE u. s. ATOMIC ENERGY COMMISSION]
The Acid Hydrolysis (Aquation) of the Trichloroammineplatinate (11) Ion BY THOMAS S.ELLEMAN, JOHN W. REISHUS AND DONS. MARTIN, JR. RECEIVED AUGUST22, 1957 The acid hydrolysis of [Pt(NHa)Cla]- in aqueous solution has been studied by spectrophotometric and potentiotnetiic titration techniques in the temperature range of 0 to 33'.
(NHa)Clz(HzO)]
+ C1-, kl is 3.6 X loT5set.-' and k-1 is 2.5 X
For the first hydrolysis: [Pt(NHa)Cla]1 ~ i i o l e s -sec.-I ~ a t 25'.
AHl
+ H20
+ 1120
+ C1-.
Approximate values, K z = kz/k-z = 4
Introduction In an earlier publication1 i t was shown that the tetrachloroplatinate(I1) ion undergoes a reversible (1) L. F. Grantham, T. S. Elleman and D. S. Marttn, Jr , Tars 2966 (1955).
JOURNAL, ??,
X
__
[Pt-
* is 18.9 kcal. andk-AH-1 *
is 16.8 kcal. The existence of a second acid hydrolysis has bceii detnonstratecl: [Pt(NHd)C1z(HzO)] (NHa)CI(Hg0)2]+ were indicated.
k- 1 ki
z
[Pt-
ks
niole/l and k-z = 0.2 1. mole-' set.-' a t 24'
acid hydrolysis reaction in aqueous solution for which rate constants and an equilibriun constant were Presented. A similar aquation has been observed for C ~ ~ - [ P ~ ( N H ~ >and Z C [Pt(en)C121, IZI (en is ethylenediamine) by Banerjea, Basolo and Pear-
Feb. 5 , 1958
ACID HYDROLYSIS OF TRICHLOROAMMINEPLATINATE(~~) ION 537
sure below 45' to avoid deposition of Pt metal. The crystals were taken up in a small volume of warm water and recrystallized in an ice-bath. They were washed finally with alcohol and ether and dried a t 60'. Typical analyses of freshly dried crystals were: K = 10.8, Pt = 54.6, C1 = 31.2, 28.4, NH3 = 5.1; calculated K = 10.9, Pt = 54.5, C1 29.6, NHI = 4.8. Analyses for chloride were made gravimetrically as AgCI, after platinum had been plated out of a basic solution. Platinum was determined gravimetrically by electroplating on Pt-electrodes from an HzSOl solution containing a trace of nitrate. The potassium and ammonium ion in the solution were determined by the use of the sodium tetraphenylboron reagent, following the procedure of Gloss.6 Analyses indicated that in the laboratory atmosphere the crystals slowly added approximately one HzO per mole. However it was shown that the HzO did not replace ligands on the platinum. The sodium tetraphenylboron was purchased from the Hach Chemical Company. It was dissolved and filtered to prepare the precipitating reagent. For ion-exchange separations, Dowex-1 anion-exchange [Pt(NI&)Cl,]Ha0 resin, screened 20/50 mesh (inches), was employed. Other ki reagents met ACS specifications. Water from the laboratory distilled water tap was redistilled from alkaline per[Pt(XHa)Clz(HrO)] f C1- (1) manganate. the replaced C1- might be either cis or trans to the Equipment.-Solutions were thcrmostated in a waterNH3. Since the replacement of C1- by NH, to bath whose temperature was controlled to f0.1'. A regive predominantly the cis- [Pt(NH3)2C12] is often frigerating coil in the bath permitted controlled temperadown to 0". cited as a classic example of the trans-effect, which tures pH measurements were made with a Beckman model "G" was reviewed by6,Quagliano and Schubert, one PH meter, with electrode model 1190-80, calibrated with a expects the cis-chloride t o be the more labile. standard buffer at pH 4 or 7. Ultraviolet spectra were scanned with a Cary recording However, the possibility exists that a mixture of the model 12. For following kinetics, the isomers may form. Also, in case of a second spectrophotometer, absorbance of a solution, A , a t a fixed wave length, was aquation followed as a function of time by a Beckman model DU spectrophotometer. The conventional definition of abk- z sorbance is used: A = log I o / I ,where I is the intensity of [Pt(iYH3)Clz(H~O)] Hz0 light passing through the solution cell and IOis the intensity k2 of light passing through solvent in an equivalent cell. [Pt(NH3)C1(Hz0)zIf f C1- (2) Behavior of K[Pt(NH3)Cla] Solutions.-It was observed when K[Pt(NH3)Cla] dissolves in HzO at room tema cis- or a trans-isomer may result. According to a that perature, the PH of the solution falls and slowly approaches summary given by Chatt, et ~ l . NH3 , ~ and H20 a value of 4.5 to 5.0. A simultaneous change in the ultrado not differ greatly in their trans-directing effects yiolet spectrum occurs, and the titer of the solution by base which are relatively weak. If the trans-directing increases. These changes ensue over a period of several and they are reversed by the addition of KCl. Such effect of NH3 is greater than that of HzO, following hours behavior is consistent with the reversible replacement of a the order given by Chatt, et al., the cis-[Pt(NH3)- chloride ligand by a water molecule if the aquo-ligand is a C1(H20)2]+would predominate even if i t were weak acid which can dissociate to give H + and [Pt(SH3)C l ~ ( o H j ] - . If an SH3 were replaced by HzO, the addition formed from the cis- [Pt(NH&%(H20) of C1- would yield [PtClal-. The presence of [PtC14]- is not detectable in the solution, although tests showed that Experimental the presence of 1% [PtC14]- gave discernible changes in Materials.-K [Pt(K'H3)C13] was prepared from the Kz- the ultraviolet spectrum. Apparently, KH3 is not replaced. [PtCla]starting material.' To attain a satisfactory purity, A number of solutions of iX[Pt(i'iH~)Cl,] in water were the final salt was crystallized from a concentrated solution prepared in which NaZSO4 was added to give an initial ionic which contained no excess chloride. An involved procedure strength of 0.318 mole/l. The solutions were allowed to was required to attain these conditions. stand in a constant temperature bath to attain a steadyThe preparation utilized suggestions of Lebedinskii and state condition; usually a period of one to seven days was Golovnya.6 KZ[PtClJ was f i s t converted to cis-[Pt(NHa)~ allowed. The temperature range was 0 to 35'. Samples Cl,] by refluxing with 570 ammonium acetate. The crys- were withdrawn, and the equivalents of acid present were tallized cis-[Pt(KH3)~Clz]was refluxed for cu. 2 hr. with determined by the procedure which was used for the study of concd. HCl in the presence of some reduced platinum metal [PtCl4]'.' A typical titration curve is shown in Fig. 1. catalyst. After the mixture was cooled, unreacted cis- In this figure the blank titer has been subtracted. If the [ P ~ ( N H J ) ~ Cand ~ Z ] catalyst were removed by filtration. degree of second aquation described by equation 2 were neg[Pt(NH3)4] Clz was added to precipitate the golden salt ligible, the equivalents of base gave the amount of [Pt[Pt(NH&] [Pt(NHa)ClJl2. After they were washed, the (NH3)Clz(HzO)]in the solution, which also was equal to the golden crystals were mixed with a freshly prepared, concen- chloride ion. Most titrations were carried out with an intrated solution of a stoichiometric quantity of Kz[PtCL] and itial platinum complex concentration of 0.0166 M ; however, heated for a short time to precipitate the LPt(NHs)4lf+ a few were made a t a concentration of 0.00415 M to test the from the golden salt as the highly insoluble Magnus green consistency of the equilibrium constant for the system. salt, [Pt(NHa),] [PtCL]. The resulting solution of K[PtThe kinetics of the aquation was conveniently followed (NHa)Cla] was evaporated to dryness under reduced pres- spectrophotometrically. The absorbance a t 343 mp of a 0.0166 M K [Pt(iXH3)C13]solution, containing Na2SOc t o (2) D. Banerjea. F. Basolo and R. G. Pearson, THISJOURNAL, 79, give an ionic strength of 0.318 mole/l., was followed with 4055 (1957). the Beckman DU spectrophotometer until the steady state (3) J. V. Quagliano and L. Schubert, Chem. Revs., 60, 201 (1952). was attained. It was assumed that the only significant (4) J. Chatt, L. A. Duncanson and L. M. Venanzi, J . Chem. Soc., platinum species were [Pt(NH3jC13]- and [Pt(NHa)C124456 (1955). ( H z ~ ) ] .The concentrations could then be calculated as (5) V. V. Lebedinskii and V. A. Golovnya, Iswest. Scklora P l o f i n y
son.2 However, they found that aquation was not observable for trans- [Pt(NH3)&] or for [Pt(NH3)aCI1'. The present paper concerns a study of the kinetics and equilibrium behavior of aqueous systems containing K[Pt(NH3)Cl3]. The aquation process, in which a chloride is reversibly replaced by H20, provides a means for the isotopic exchange of chloride ligands with chloride ion in the solution. This work was undertaken to supplement a study of such isotopic exchange which is t o be reported later. With the [Pt(NH3)C&]- system special consideration needs to be given t o the possible nonequivalence of the chlorides in the well-recognized square planar complexes. Thus in the reaction k- 1
__
+
+
I.
i Drrhg. Blagorod. A4etal I n s l . Obshchci i h'eorg. K h i m . A k a d . Na74k s.s.s.R., ao, 95 (1947).
(6) G . H. Gloss, Chemist Analyst (Baker Chem. Co.),43, 50 (1953).
T. S.ELLEMAN, J. W. REISHUS AND D.
538
s. I\/IARTIN, JR.
Vol. 80
After they had aged for several days, the effluent solutions from the anion-exchange columns were chemically analyzed for platinum content, total equivalents of acid and total chloride. Results of the analyses are in Table I .
TABLE I AAALYSES OF SoLwrIox SEPARATED BY ANION EXCHANGE FROM ACEDK[Pt(NHa)Cl3] S O L U r I O N S Effluent a t room temperature, 24 &lo Concn , mmoles./ml.
Constituent
2.40x 10--3
T(Itd1 P t Total Cl Equiv. acid
0
I
2
MI. N a O H
3 ADDED.
4
5
Fig. 1.-Titration curve of ail aged solution of 0.0166Jf K[Pt(NHa)Cla]; TazSOl added t o give ionic strength of 0.318 mole/l.
a function of time from the absorbance of the initial K[Pt(SH3)Clal solution and from the absorbance of the equilibrium state. [Pt(NH3)CIz(H20)] Solutions.-Since [Pt(NH3)CIB(HzO)] species is uncharged, it was possible to prepare solutions of this compound which were relatively free from [Pt(XHs)Cl,] - and C1- by the use of an anion-exchange resin. The formation of the aquo-species was favored by dilution, so approximately molar solutions of K [Pt(NH3)Cl3] were aged t o establish the steady state. This solution was then passed at a rate of about 5 ml./inin. through a 1.5 cm. X 40 cm. column filled with Dowes-1 anion-exchange resin in the sulfate form. The ultraviolet spectrum of such an effluent solution is shown in Fig. 2 which also includes the spectrum - for comparison. of [Pt(SH3)Cl3]
4.i3 X
2.71
x
10-3
Base Hydrolysis of [Pt(NH3)CI2OH]-.-Since cis- [Pt(NH3)zCIJ is prepared by the action of ammonium acetate on [PtCL]-, stoichiometric amounts of ammonium acetate were added t o dilute solutions of the anion-column separated [Pt(SHa)CL(H20)]. To other solutions were also added NH3, KOH and K(CzH302) in amounts equivalent t o the platinum complex. A dark-green precipitate usually was observed after about 1 hr. with 0.01 M platinum concentrations. S o precipitation was discerned over a period of several hours with 0.001 M concentration, and changes in the ultraviolet spectrum could be followed readily for each of the bases. All of the reagents appeared t o produce eventually very similar spectra. In Fig. 3 is given an exI
I
I
320
WAVE
LENGTH
1
I
280
240
200
mp
Fig. 3.-Effect of the addition of NH3 on the ultraviolet spectrum of aged [Pt(NH3)C12(HBO)] solution to give 0.001 111 total Pt and 0.0015 M S H 3 ; temp., 24 =tl o , 10.00 cm. cells: (1) Initial [Pt(NH3)Clz(HzO)]soh.; (2) 2 min. (NIh added a t t = 0 ) ; (3) 17 min.; (4) 53 min.; (5) 85 miti.; (6) 154 miti.
04+L
ample of the observed spectral changes when NH3 was added. Furthermore, analysis of the solution showed that uncomplexed NH3had not changed significantly. These results were interpreted t o indicate that a second hydrolysis Fig. 2.-Ultraviolet spectra of solutions: [Pt(XHa)C13]-, had ciccurred, giving the complex [Pt(NH3)Cl(OH)s]-. spectrum for 0.001 ;IT solution of K[Pt(NH3)Clsl in 10.00cm. The spectral changes, shown in Fig. 4, followed after the cell with excess KC1 t o supprcss aquation; [Pt(XH,)Cl,- amiiivniacal hydrolysis solution had been treated with H B (HzO)], spectrum for 0.001 df solution in 10.00 cm. cell. SO4 and KC1. I t can be seen that the rcaction had been reThis solution is an aged effluent from the Dowex-1 nilion- versed, and it appears from the spectra that [Pt(SH3)Cls(HzO)] re-forms. In other cases for which much higher exchange colunlns in which [ P t ( S H 3 ) C 1 ~ ( I I ~ 0is) ]the dom- concentrations of chloride were added, the final spectrum inant species, pH 4.0; [Pt(SH31C12(OH)] -, spectrum ob- changes indicated first the formation of [Pt(NHa')Cl?tained when the pHof the solution For the [Pt(SHa)Cls(HzO)] (HzO)] followed by nea-11- complete conversion to [Pt(hHa'lCla] -. spectrum was raised to 9.0 by the addition of S a O H . 400
360
320
280
240
W A V E LENGTH m p .
The titration curve of Fig. 1 indicates that [Pt(NH:jjCln(H20)]is an acid with a pK of about 7. Seutralization of the free acid by base gave inirnediately the third spectrum shown in Fig. 2, labelled [Pt(XH~)Clz(OH)]-since this anion must be the predominant platinum species. .llthough thcre were distinct differences, the spectrum of [Pt(SHaiC1,[OH'~]was uot greatly different from that c i f its conjugate acid. After the neutralization, however, slow changes ill the spectrum iiitlic,ttetl that :L furtlicr SUI,stitution reactioii occurred.
Results and Discussion Acid Hydrolysis Equilibria.-The system was considered in which R [Pt(XH3)C13]and KCI were dissolved in H?O a n d allowed t o approach a steady state with regard t o the reactions 1 and 2. If I
