49ti8
HANSU .
JONASSEN,
K. BRUCE LEBLANC AND RUTHM. KOGAN
Vol. 72
butrers alkaline to the isoelectric point consistently presented a single boundary, At pH 6.5, no heterogeneity was observed in acetate or phosphate buffer a t ionic strength 0.1 after electrophoresis for three or five hours. The curves relating mobility with pH for 81- and normal 8lactoglobulin intersect a t about pH ti.6 (Fig. 3), so that if heterogeneity were to be observed a t an alkaline PH it would seem least likely to appear a t pH 6.5 and most likely to appear a t pH ,5.5 or S.1. The nature of the heterogeneity presented by $-lactoglobulin has been the subject of much conjecture.’* I t apparently is not the result of the isolation procedure. Crystals obtained by alcohol fractionation, as well as by salt fractionation of whey; crystals from the milk of a single animal, as well as mixed commercial milk; and even crystals from normal pasteurized milk, showed the same consistent electrophoretic heterogeneity a t pH 4.8 in acetate buffer a t ionic strength 0.1. The only exceptions to this highly reproducible pattern were found in the crystalline lactoglobulin isolated from the first-day colostrum of one cow and from the milk of a cow infected with brucellosis. I n both cases, the pl-lactoglobulin predominated, amounting to SOY; in the brucellosis @-lactoglobulin and lOO?; in the colostrum. The possibility of aggregation or dissociation of the molecule is discredited by finding similar molecular weights for the normal and dl-lactoglobulins. The isolation of one of the components in pure crystalline form that migrates with the velocity of the 8-lactoglobulin in alkaline buffer, although i t has a mobility similar to that of the immune globulins of whey a t pH 4.S, refutes the postulate by Lundgren and Ward that the heterogeneity of P-lactoglobulin is d result of the contamination with one or more of the immune globulins characterized by Smith l q
In view of the hiih molecular weight (about 1X0,OOO) and the presence of carbohydrate and phosphorus in these immune globulins, it is impossible for an immune globulin to account for any component that constitutes the elec trophoretic heterogeneity reported here. The heterogeneity of P-lactoglobulin manifest a t pH values acid to the isolectric point can be ascribed primarily to :t basic difference in the net charge. This may be attributed to a fundamental variation in the structure of the protein molecule or to the conjugation of the same molecule with another component, with resultant change in the dissociation of ionizable groups and hence difference in the net charge. Acknowledgment.-The authors are greatly indebted to C. A. Zittle for the many helpful discussions and criticisms of this paper, and to E. S. DellaMonica for assistance in preparing the normal crystalline 8-lactoglobulin.
(18) Liindgren aud \Yard, .4nti Ret H i o i i i r i n , 18, 111 (1‘44Q) ,IO) Smith, T A i d CLr!!8 166, 665 l l l t ~ i ~
PHILADELPHIA 18, PEUNSYLVANIA RECEIVED FEBRUARY 3, 1950
Summary Crystalline /?-lactoglobulin, presenting a single electrophoretic boundary in buffers of ionic strength 0.1, alkaline to the isoelectric point, has been found heterogeneous in buffers acid to the isoelectric point. By means of alcohol fractionation and differential solubility a t pH 4.8 and 5.3, one of the two components resolved by electrophoresis in acetate buffer a t pH 4.8 has been isolated as a crystalline protein showing a single electrophoretic boundary over the conventional pH range in buffere of ionic strength 0.1. The per cent. nitrogen, molecular weight, optical rotation, optical density (2801250 mF), biuret color density, solubility in water and salt, electrophoretic mobility and isoelectric point of the single-boundaried component and of the normal complex are compared.
[~OPITTRIBUTKOK FROM THE KICHARDSOS CHEMICAL L A R O R A ORIES ~ O F THE
TITLASE ITVIVERSITY
OF
LOUISIANA 1
Inorganic Complex Compounds Containing Polydentate Groups. IV. Formation Constants of Diethylenetriamine-Nickel(I1) and --Copper(I1) Complexes 13p H.im
H. JONASSEX, li. RKUCELE:BI..\KCA N D RUTHXI. R O G ~
If a iiietal iou cat1 coordinate two or inore groups in the formation of complex ions, it will do so stepwise, and intermediate equilibria will be present. If M represents the metal ion and -4 the coordinating group the equilibria are
f o r the individual steps is the over-all coniplexity
constant for the equilibrium XI
+ -YA
MA,?; k v
=
[MA~l/’[MI[Al”
where LV is the maximum number of groups coordinated. M +A MA; kl = [;MA]/[M][A] Bjerruml devised a method for measuring these step equilibria constants (or formation conMA + A JJ MAS; k2 = iMAzl/[MA][Al XIA,. I 4-A 31.4,; k,, ~ ~ l ‘ ~ , ’ ~ , ’ [ ~ I . 4 ~ - 1 ~stants) 1 ~ 4 1 for basic coordinating groups by measuring the pH of solutions which contain known amounts Each of these equilibria cat1 be represented by (1) J. Bierrum, “Metal Ammine Formation in Aqueous Solution,’ I’ Iiaasr and Son, Copenhrgen, 1‘111 :L constant, and the product of all the constants
FORMATION CONSTANTS OF DIETHYLENETRIAMINE-NICKEL(II) 4969
Nov., 1950
of the complexing agent and the metal ion. This method holds for polydentate groups only if the coordinating group is completely bound by chelation to the metal ion and exerts no basic function. This paper reports the study of the formation of complexes between diethylenetriamine (abbreviated dien) and the copper(I1) and nickel(I1) ions. Calculation of Constants The following symbols will be used CHNO,= total concn. of HNOa in the solution
CM,
= total concn. of metal ion in the solution = total concn. of dien in the solution
C'dien = total concn. of dien not complexed by the metal ion in the solution C, = total concn. of hydrogen ions bound to the uncomdexed dien C. = CHNO~'- [ H + ] [OH-] = [dienH+l 2 [dienHz+ 2 ] 3 [dienH3+a]
++
where '%&en is the mean number of hydrogen ions bound by each dien molecule and kl,kz and k3 refer to the stepwise dissociation constants of dienHa+3. = [dien]/C'dien
(2)
where Ql&n refers to the fraction of the dien in the acid-base system which exists as the free dien. The following expression can be derived from ( 2 ) and the equations for the acid-base constants of dien Defining E as the mean number of dien molecules attached to each metal ion, the following equation is obtained
-n =
(Cdien
- C'dien)/CMs
A combination of (1) and (2) gives [dienl =
(Ydiencahdien
or pldienl = log (zdienloldien)
Discussion The authors2 previously measured the acidbase pK of dien a t 30" and 40" a t the same salt concentration as above Pkl fik2 Pks
+
(1)
(Ydien
M . It has been shown3 that in these concentration ranges of chloride ions no appreciable amounts of chlorocomplexes are formed. Individual solutions of approximate constant ionic strength were prepared for each measurement. These solutions were 0.1 M in metal ion, 0.1 M in "4, and of varying concentration in dien. The sum of the concentration of potassium chloride and potassium nitrate was 1 .oo M . The pH values of the solutions were measured with a Beckman Model G PH meter standardized with Beckman buffers, pH of 4, 7 and 10. The solutions were maintained a t 30 and 40" in constant temperature baths.
- log CS
(4) (5)
If the acid-base constants of dienj2 the concentration of the various components of the solution, and the p H of the solution are known, Z and p[dien] can be calculated from equations (1)) ( 2 ) )(31, (4) and (5). Experimental
30'
40'
4.78 9.23 9.94
4.59 8.94 9.68
With the use of these acid-base constants, the quantities Z and p [dienJ were calculated from equations (4) and ( 5 ) . Table I contains data on the copper(I1) ion and dien at 30". TABLE I PH MEASUREMENTS O F c O P P E R ( I 1 ) I O N SOLUT~ONS CONTAINING DIETHYLENETRIAMINE AT 30" CC~(I= I ) 0.1055, C K N O4~ CKOI= 1.00 M -log ltdien cs ?2 p[dien] C'dien cdidien 9H Udisn zdien 0.0350 ,0401 .0452 .0500 .0648 ,0749 ,0850 IO900 ,1000 ,1054 ,1199 .1403 .1599 ,1799 .2001 ,2198 ,2399 ,2600 ,2800
2.53 2.70 2.79 2.84 2.98 3.05 3.10 3.13 3.19 3.21 3.37 4.19 4.96 6.89 7.56 8.38 8.82 9.17 9.40
16.84 16.33 16.06 15.91 15.49 15.28 15.13 15.03 14.86 14.80 14.33 11.92 9.85 5.70 4.36 2.74 1.91 1.29 0.91
3 . 0 0 0.0970 0.0324 ,0982 ,0328 2.99 ,0986 ,0330 2.99 ,0986 ,0330 2.99 ,0988 ,0331 2.99 .OM9 ,0332 2.98 ,0992 ,0333 2.98 ,0994 0334 2.98 ,0334 2.98 .0994 ,0997 2.97 ,0335 .Oil93 ,0335 2.96 ,1005 2.80 ,0359 .lo02 ,0418 2.40 2.00 .loo0 ,0499 .lo04 ,0508 1.98 ,0999 1.87 ,0535 ,1000 1.69 ,0593 1.42 ,1006 ,0708 ,0838 ,1005 1.20
0.025 .069 ,116 ,161 .300 ,395 .490 ,536 ,631 ,682 ,819 ,990 1.044 1.232 1.416 1.579 1.711 1.793 1.860
17.85 17.34 17.07 16.92 16.50 16.28 16.13 16.03 l5,86 15.80 15.33 12.92 10.85 6.70 5.36 3.74 2.91 2.29 1.91
Diethylenetriamine was purified by the method previously described.2 This consisted of vacuum distillation and precipitation of its monohydrochloride from alcohol. A stock solution, approximately 1 molar in concentration, was exactly neutralized with potassium hydroxide and standardized potentiometrically against standard acid. The neutralization with potassium hydroxide produced an equimolar concentration of potassium chloride in the stock solution of dien. The concentration of chloride ions in the solutions from which the log k values were determined did not exceed 0.2
Similar data, obtained a t 4.0") are omitted. After Fz exceeds 1, the values of P[dien] decrease very rapidly as Z increases. This gives a flattening of the curve as can be seen in Fig. 1, which shows the formation curves a t 30 and 40". The curve does not approach an Z value of 2 asymptotically, as does the curve of a typical B-coordinated metal ion. This indicates that the copper (11) ion does not exhibit a stable coordination number of six in aqueous solution a t this concentration. Obviously, a second dien molecule is complexed by the copper(I1) ion, but i t is not coordinated through all three of its amino groups.
(2) H. B. Jonassen, R. B. LeBlanc, A. W. Meibohm and R . M . Rogan, THISJOURNAL, 71, 2430 (1950).
(3) G . A. Carlson, J. P. McReynolds and Frank H. Verhoek, ibid., 67, 1334 (1945).
HANSB.
4970
JONASSEN.R.
BRUCELEBLANC AND
X coordination nuniber less than six for the copper(I1) ion is indicated.
11
7
j
1
13
13
RUTH
Vol. 72
i'd. ROGAN
Experimental points on the curve between 2 values of 0.3 and 0.7 yield the values of log kl, listed in Table 11, for the copper(I1) ion a t 30" and 40". From the data in Table I1 the heat of binding of [Cu(dien)3 f 2 is calculated from
and found to be -21 kcal. The data for the formation of the nickel(I1) complex ions and &en a t 30" and 40" were obtained in a manner similar to that for the copper(I1) ions. A plot of these data is shown in Fig. '7.
1;
p [dien] Fig. 1 -Formation
curve of the copper(I1) ion and t i l ethylenetriamine d t 30 and 40' 0, S O " , X , 40c
These data are not a t variance with the observations on the copper (11) ethylenediamine system reported by B j e r r u n ~ ,who ~ found a maximum coordination number of 3 for the copper(I1) ethylenediamine complexes. The polarographic investigations on the copper (111 diethylene triamine systetn5 also indicate that, in the presence of a large excess of dien, two molecules of dien are coordinated to the copper(I1) ion. S o calculations of individual formation constants of the complex can be made from polarographic data. Because of the assumption in deriving the equations, the coordination of the second dien through only one or two of its amino groups makes the equations inapplicable for calculation of the second formation constant of the copper(J1) ion. The logarithm of the first formation constant can be calculated from log ki = p[dien] i
+ log ( r z / ( l
\HIL
-
(6 1
11)
I1
FORMATIIOU CurYsrarrs OF THE COPPER(^^) Iua DIETHYLENETRIAMINE AT 30 AND 40" $[dien]
ii
log i z / ( l
- Z)
A\U
log k t
1 = 30"
0.300 395
16 lti lti lti 1.5
,490 5% .(531 .(j82
-0.37 - 18 - 02 r 00
50 28 13
0.3
+
86 15 811
+
16 16 16 16
Id 10 11 09
2.$
Ili 09
3.3
1h 13 16 11
A\.
16 O d 15 6ti 15 55 15 4b 15.43 15 28
490
536 584 ,631 681
--
'.
i- _ _ _ ___-2
-0 37 02 -i-.06 .I5 .23 4- 33
-
+ +
1,; tifj 15 b4 15 61 15 61 15.66 15 61 .I\I:, o:?
(4) J Bjerrurn and 0 J ilrlsen, A r t Lhcm Scaitd., I,297 (1048) ( 5 ) H A I.aitinen, B I Omtott, J C B d a r , Jr , and Sherlock 551 inn Ji ?rus JOT r
