Feb. 5, 1959
HYDROLYSIS AND DIMERIZATION O F COPPER(I1) CHELATES O F 1,2-DIAMINES [CONTRIBUTION OF
THE
525
DEPARTMENT O F CHEMISTRY OF CLARK UNIVERSITY]
Hydrolytic Tendencies of Metal Chelate Compounds. V. Hydrolysis and Dimerization of Copper(I1) Chelates of 1 +Diamines' BY R. L. GUSTAFSON AND A. E. MARTELL RECEIVED JULY 18, 1958 Stability constants of the 1: 1 Cu(11) chelates of N,N'-dimethylethylenediamine, N,N,N',N'-tetramethylethylenediamine, N-hydroxyethylethylenediamine, N,N'-dihydroxyethylethylenediamine, a,cu'-dipyridyl and 1,lO-phenanthroline are reported. The first and second hydrolysis constants of these chelate compounds are determined by potentiometric measurements of hydrogen ion concentration at 0.3, 25.0 and 42.5". In all cases the monohydroxo chelate compounds were found to be in equilibrium with binuclear diolated forms. The equilibrium constants and the thermodynamic changes involved in hydrolysis and olation reactions are discussed in the light of the structures and other properties of the ligands and of the chelate compounds formed.
I n a recent publication Courtney, et aLI2have hydroxo species, Cu[OH]zA and a dimer, reported equilibrium constants of the hydrolytic ( C U [ O H ] A ) ~ ~ +The . solution equilibria may be reactions of copper(I1) chelates of five aliphatic expressed in terms of the equations polyamines. I n the present paper the scope of this C ~ ( H z 0 ) f ' A 1_ C U A ( H ~ O ) ~ ~ 2H20 + research is extended to include four additional bidentate ligands : N,N'-dimethylethylenediamine (DMEN), N,N,N',N'-tetramethylethylenediamine (TMEN), ala'-dipyridyl (DIPY) and 1,lO-phenanthroline (PHEN), and measurements are reported a t temperatures of 0.3, 25.0 and 42.5'. Also, the work previously reported2 for the copper(I1) chelates of N-hydroxyethylethylenediamine [CU[OH]1-41[ H 1' (HEN) and N,N'-dihydroxyethylethylenediamine Kvro~lvr= (3) [CUA] (2-HEN) a t 25.0' has been extended to include data a t the other two temperatures. Experimental
+
Materials.-Samples of N-hydroxyethylethylenediamine and N,N'-dihydroxyethylethylenediamine obtained from Carbon and Carbide Chemical Co. were purified by fractional distillation and were then isolated as the dihydrochlorides. Samples of the dihydrochlorides of N,N'-dimethylethylenediamine and N,N,N',N'-tetramethylethylenediamine were prepared and recrystallized by conventional methods. The 1,lO-phenanthroline monohydrochloride was obtained from the G. Frederick Smith Chemical Co. Standardization of the aqueous solutions of the above chelating agents was carried out by means of potentiometric titration with standard carbonate;free potassium hydroxide. A reagent grade sample of a,a,-dipyridyl was obtained from the Fluka Chemical Co., Switzerland, and was used without further purification. Potentiometric Titrations.-The calculation of the distribution of the various chelate species as a function of pH and of total metal ion concentration was carried out on the basis of potentiometric titrations of the metal chelate systems under investigation over a tenfold range of concentration in the manner previously described.2 The hydrogen ion concentration was recorded with a Beckman Model G pH meter fitted with extension glass and calomel electrodes. The titrations were carried out in a multi-necked flask designed to accommodate a mercury seal stirrer, gas inlet and outlet tubes, microburet delivery tube and electrodes. The ionic strength was maintained relatively constant by using an electrolyte medium of 0.1 M potassium nitrate, and presaturated nitrogen was bubbled through the solutions during the course of the titrations to exclude carbon dioxide. Measurements were made a t temperatures of 0.3, 25.0 and 42.5'.
Mathematical Treatment of Data I n all of the copper (11) chelates studied in this investigation, four species have been shown to exist: a diaquo chelate, CUA(HZO)~~+, a monohydroxo compoundIS Cu [OH]A(H20)l + , a dihy(1) This investigation was supported by a research grant, B-1171, from the National Institute of Neurological Diseases and Blindness, Public Health Service. (2) R. C. Courtney, R. L. Gustafson, S. Chaberek, Jr., and A. E. Martell, THISJOURNAL, 81, 619 (1959).
+
~ C U [ O H ] A ( H ~ O )JJ ' + (Cu[OH]A)e2+
+ 2Ho0 (5)
where A represents one of the ligands listed above. The amount and distribution of the various chelate species present under varying conditions of pH and total concentration may be calculated from the above equilibria with the relationships outlined in a previous paper.2 Since the present data are to be correlated with kinetic data obtained from the Cu(I1) chelate catalyzed hydrolysis of substituted phosphonofluoridates, it is necessary t o consider the unbound Cu(I1) ion as a significant constituent of the experimental solution. The metal chelate formation constant KMAmay be calculated from the expression
where (7)
for the chelates of DMEN, TMEN, H E N and 2H E N and (7')
for the chelates of DIPY and PHEN. TA is the total concentration of ligand species present, and K , and Kz are the first and second acid dissociation (3) Brackets around OH within the formula of a coijrdination compound indicate a hydroxyl ion bound directly to the metal ion, whereas a set of brackets enclosing the entire formula indicates molar concentration.
R.L.GUSTAFSON AND A. E. MARTELL
526
Vol. 81
pH values in the m interval from 2-3, where m represents the number of moles of potassium hydroxide added per mole of metal ion. The portion of the curves corresponding to m values from 0 to 2 (2 - ~ ) T A [H+] 4- [OH-] [A] = (8) indicates neutralization of the two moles of HCl 2iH+12 . IH+1 +present in the molecule. Beyond the crossover of KiKz KP the curves a t m = 3, the second buffer region is diswhere a is equal to the number of moles of standard placed to higher p H values with a corresponding KOH added per mole of ligand. With equations 6-8 increase in the concentration of the metal chelate. i t was possible to calculate formation constants Similar sets of titration curves were obtained for for the copper(I1) chelates of DMEN, TMEN, the copper(I1) chelates of the other ligands studH E N and 2-HEN. However since the copper(I1) ied. The greater requirement of base a t a particuchelates of DIPY and PHEN are virtually com- lar p H as a result of the higher concentration of pletely formed even in acid solution, it was not metal chelate is an indication of the formation of a in this manner. For a polymer of the type (Cu[OH]A,"+). As is shown possible to calculate KM\?A solution containing two moles of the acid form graphically in Fig. 2, a plot [H+1 (TOE [H+] of the ligand per mole of copper(I1) ion, the fol- [OH-])/ [CUA] V S . [CuA]/[H+] results in a straight lowing equation applies line, which verifies the assumption of a dimer of KMA= the Cu(I1)-TMEN chelate compound. Similar Y plots were obtained for the other metal chelate sys(9) K M A ~ [ ~ ~] ~~( T TX -A [H'I f [OH-] f [AI f u) tems investigated, but in some cases of dipyridylCu(I1) and of o-phenanthroline-Cu(II), the interwhere cepts a t [CuA]/[H+] = 0 fell a t values which were actually slightly below zero. For the other systems studied, intercepts a t positive values were and K M A ,is expressed by the equation obtained, indicating extensive formation of the mononuclear monohydroxo chelates. Obviously Cui%'+ + A CUA4z2+ some finite concentration, however small, of monohydroxo chelate species must exist in accordance with the equilibrium expression of equation 5 . From two sets of experimental data, it is possible Although it is difficult to assign accurate values of to calculate K M Aand KMA,from equation 9. KM[OH]A for Cu-DIPY and Cu-PHEN in the I n calculating the species distribution as a func- cases where a negative value of the intercept was tion of -log[H+] and total metal species con- obtained, maximum values have been chosen based centration, simultaneous solution of the equilib- on the fact that intercepts below zero have no rium expressions 1-4 is time consuming and cumbersome; hence the following procedure was emTABLE I ployed. With an estimated value of [CuA], the EQUILIBRIUM CONSTANTS O F COPPER(I1) CHELATES O F 1,2concentrations of the other chelate species were DIAMINES determined a t the desired hydrogen ion concentralog Ligand K,, pKXIOH]A ~ K M I O H ] l A pK(MIOH]A)s k: tion. The concentration of unbound copper(I1) was calculated from the relationship t = 0.3" [ CuA]X 8.54 19.7 13.41 3.7 DMEh' .. [CUI = constants of the ligand, respectively. The concentration of the neutral species of the ligand [A] may be calculated by the relationship L
1
+
(-)"I
By a method of successive approximations, such calculations were repeated until the total concentration of all species became equal to the total concentration of copper(I1) compounds T Min accordance with the equation TM = [CUI
+ [CuA] + [CU[OH].~]+ [CU[OH]ZA]42 I( CU[OHIA)zl
(13)
Values of AH, A F and A S for the hydrolysis and olation reactions were calculated with the usual relationships 2.303RTiTz (log Ke Tz - TI A F = - 2.303 R T log K A S = ~AH - A- F -
AH =
- log KI)
T
Results and Discussion Solution Equilibria.-The titration curves illustrated in Fig. 1 for the Cu(I1)-TMEN chelate show that an increase in metal chelate concentration results in a shift of the buffer regions to lower
TMEN HEN 2-HEN DIPY PHEN
..
..
..
.. ..
8.34 7.69 7.64 8.3 8.3
19.0 18.1 17.43 18.7
. . .a
13.26 13.7 13.2 11.65 11.65
3.4 1.7 2.1 5.0 4.9
t = 25.0'
12.41 3.8 18.2 17.72 12.13 3.9 12.4 2.2 17.24 12.9 1.4 16.52 10.81 5.0 17.67 17.3 10.67 >5.0 t = 42.5' 11.87 3.8 18.1 7.83 DMEN .. 11.48 3.8 7.64 16.93 TMEN .. 12.0 2.2 7.08 16.53 HEN .. 12.0 1.7 6.88 15.84 2-HEN .. 10.27 >4.7 >7.5 16.91 DIPY .. 10.32 >4.3 >7.3 17.13 PHEN .. 0 Precipitation presumably of the neutral chelate species, Cu[OH]*A, occurred in the upper buffer region in all cases where the total metal chelate concentration was greater than 3 x 10-4 M .
DMEN TMEN HEN 2-HEN DIPY PHEN
9.72 7.20 9.90 9.68 6.33 7.4
8.09 8.00 7.30 7.15 7.9 >7.8
HYDROLYSIS AND DIMERIZATION O F COPPER(I1) CHELATES O F 1,2-DIAMINES
Feb. 5, 1959
12
81
\
527
\
-
4
I
I
I
I
I
6
5
7
8
9
IO
-LOG [H'I, Fig. 1.-Potentiometric titration of solutions of copper(I1) iY,N,N',N'-tetramethylethylenediamine chelates. TA/TM = 1. Concentrations: solid line, 9.88 X M; dashed M; dotted line, 3.07 X M; dot-dash line, 5.12 X M; t = 25.0'; p = 0.10 (KNOs). line, 1.02 X
Fig. 3.-Graph showing distribution of copper( 11)N,N,N',N'-tetramethylethylenediamine chelate as a function of the hydrogen ion concentration: A, diaquo chelate, [CuAa+]: €3, monohydroxo chelate, [Cu[OH]A1+]; C, di; hydroxo chelate, [Cu[OH]pAaJ; D, dimer, [(CU[OH]A)Z~+] E, free copper(I1); OH, hydroxyl ion. t = 25.0'; p = 0.10 ( KNOd.
physical meaning and the assumption that the negative intercept is a consequence of the order of probable error in determining KM[OHIA by extrapolation. The equilibrium constants for all of the systems investigated a t temperatures of 0.3, 25.0 and 42.5' are listed in Table I. The most striking result of the equilibrium studies is the relatively high dimerization tendencies in the cases of the copper(I1) chelates of dipyridyl and o-phenanthroline. By comparing the values of the dimerization constants Kd in Table I with the values of the acid dissociation constants of the various ligands listed in Table 11, i t may be seen TABLE I1 ACID DISSOCIATION CONSTANTS OF LIGANDS t = 25.0': p = 0.10 (KN03)
O
0
2
4 6 CCuAl/ CH+l x IO-'
h
Fig. 2.-Plot of data of Fig. 1illustrating presence of a dimer of copper( 11)-N,N,N',N'-tetramethylethylenediamine.
Ligand
9Ki
9Kr
DMEN TMEX HEN 2-HEN DIPY PHEN
7.01 5.90 6.49 6.26 4.45 4.95
9.88 9.14 9.52 9.24
.. ..
that the chelates of dipyridyl and o-phenanthroline, which are by far the least basic and hence the weakest donors of the six ligands investigated, dimerize to the greatest extent. It would perhaps be expected that in a chelate compound in which the
R.L. GUSTAFSON AND A. E. MARTELL
528
place more easily in the case of the aromatic compounds. Whereas the stabilities of the aquo chelates of DMEN and T M E N are markedly different, these comDounds are observed to underpo dimerization to approximately the same degree >nd to a much greater extent than the chelates of HEN and 2HEN. Examination of the first and second dissociation constants shows that the hydrolytic tendency of Cu-TMEN is considerably greater than that of Cu-DMEN. T h i s difference would be
TABLE I11
l
v
~OF AHO(KCAL./MOLE) ~ 2 ~ ~ ~FOR HYDROLYSIS AND DIMERIZATIOS REACTIOXS OF Cu(I1) CHELATES
Ligand
AHC(KM[oE]A)
DMEN
+6.6 i0 . 2 +7 i1 $5.7 i 0 . 2 +7.1 i 0 . 3 +
