CONSTANTS OF REACTION BETWEEN SULFITE AND OXIDIZEDGLUTATHIONE 2057
April 20, 1955
The frontier electron density distribution parallels the observed order of reactivity toward electrophilic attack, namely, meso > 1 > 2 which the n-valence electron density fails to do, unless one omits from consideration the high density a t the non-perimeter atoms. Yet the ease with which the n-valence electrons are brought to the position of electrophilic attack varies, according to the implications of equation 1, in the order meso > 1 > 2 > 13 for which the self polarizabilities5 are ng,g = 0.526, = 0.454, n2,2= 0.411 and n13,lS = 0.331. I n general in alternant hydrocarbons the success of the frontier electron density distribution in reflecting the experimental data is paralleled by the failure of the a-valence electron density to do likewise, unless one excludes from consideration the large densities a t the non-perimeter atoms as in anthracene. It is clear, therefore, that considerable care must be exercised in the deduction of physical interpretations from correlations between experimental data and static n-electron configurations. It is vital to relate static configurations to changes in configuration, an investigation which advances strong support for equation 1 in preference to the frontier electron concept. Equation 1 also illustrates that in a polarization process the ease of provision of a pair of electrons a t the position of attack depends not upon the density of this pair in the unperturbed molecule, but rather upon the total density q,, or better still, upon the total nelectron configuration. Thus the process concerns both the appearance of a pair a t the position of attack and the simultaneous removal of all other a-electron pairs, two complementary processes (5) H. C. Longuet-Higgins and C. A. Coulson, J . Chcm. Soc., 971
which cannot be separated. Hence in the unperturbed ground state qr and nr,rare the appropriate criteria and not the density of a particular pair. Finally the details of the polarization process provide a link between the two methods of approximation used for the description of the reactions of conjugated molecules, namely, the method of equation 1 and the alternative transition state method of Wheland.6 The numerical correlation observed hitherto is now consolidated by the obvious physical connection indicated by the change in electron configuration implicit in (1) toward that used in Wheland's method. Numerically the n-valence electron density in benzene, for example, increases at the position of attack from 0.333 in the ground state to 1,378 for the change in coulomb integral Bar = 20, a change which has been used to represent the effect of the neutral oxygen atom in a conjugated m ~ l e c u l and e ~ ~which, ~ therefore, does not appear to be excessive as a representation of the effect due to the net positive charge carried by a neighboring electrophilic reagent. Thus the n-valence electron density achieves a high degree of localization a t the position of attack for a relatively small perturbation. Nucleophilic reactions can be considered in exactly the same way as for electrophilic. The details of the charge shifts are less interesting since the whole system of n-electrons is removed from the position of attack, as in an analogous classical problem. Nevertheless the charge shifts take place in accordance with the physical and quantum mechanical principles enumerated earlier. (6) G. W. Wheland, THIS JOURNAL, 64, 900 (1942). (7) G. W. Wheland and L. Pauling, i b i d . , 67, 2086 (1935)
ALDGATE, LONDON, ENGLAND
(1949).
[CONTRIBUTIONFROM
THE
SCHOOL OF CHEMISTRY OF
THE
UNIVERSITY OF MINNESOTA]
Equilibrium Constants of the Reaction between Sulfite and Oxidized Glutathione BY W. STRICKS, I. M. KOLTHOFF AND R.
c. KAPOOR
RECEIVED OCTOBER 16, 1954
+
+
The reversible reactions between sulfite or bisulfite with oxidized glutathione (GSSG 4-SOa' F! GSGSSOa-, GSSG GSSOa-) have been studied quantitatively in the PH range 12.4 to 5.2. From the results, four equilibrium constants for the reactions involving SOS' and the varioils charge types of oxidized and reduced glutathione and glutathione sulfonate and one constant for the reaction of one charge type with H,SOa- have been calculated. The heat of reaction was estimated from the values of the equilibrium constants a t 12 and 25 . The results were compared with data for the analogous reactions between sulfite with cystine and with dithiodiglycolic acid.
HSOs- F? GSH
+
In the course of investigations on the reactivity of disulfide groups in proteins and peptides i t appeared Of interest to study in a quantitative way the reaction between oxidized glutathione and SUIfite and to compare the results with the data previously reported' for the analogous reactions of SUIfite and cystine and dithiodiglycolic acid. The reaction between oxidized glutathione (GSSG) and sulfite was studied a t various p H and temperatures. Equilibrium constantsand the heat of reaction have been calculated from results obtained by the polarographic method of analysis. (I) W. Stricks aod I. M . Kolthoff, T H IJOURNAL, ~ IS, 4569 (1951).
Materials.-Oxidized glutathione was prepared from glutathime in the reduced state by a method described in a previous paper.s Sodium sulfite was a C.P.reagent grade Merck product. All the other chemicals were C.P.reagent grade products. Stock solutions of oxidized glutathione were 0.02 M in GSSG. The stock solutions were analyzed for GSSG by the polarographic method2 and also by amperometric mercurimetric titration at the dropping mercury electrode as indicator electrode.' Stock solutions of sulfite were prepared with air-free water and were 0.2 t o 0.5 M in sodium sulfite. They were analyzed for sulfite iodometrically. Only freshly prepared sulfite solutions were used. (2) W. Stricks and I. M. Kolthoff, ibid., 74, 4646 (1952). (3) W. Stricks, I. M. Kolthoff and N. Tanaka, Anal. Chcm., 96, 299 (1964).
w. STRICKS, I. M. KOLTHOFFAND R. c. KAPOOR
2058
The composition of the stock solutions used for the preparation of the buffers was essentially identical with that given pre\.iously.I
L'ol. 77
Results and Discussion
The polarogram of an equilibrium mixture composed of 2 ml. of 0.5 M sodium sulfite solution and Experimental 18 ml. of 2.2 X M GSSG solution in an ammoCtirrerit voltage cur\'es were measured a t 12 and 25' with nia buffer of PH 9.23 (2.5 X Af in thymol) is the manual apparatus and circuit described by Lingane and Kdthoff' and automatically with a Heyrovsky t q e self- shown in Fig. 1A. Thymol, which has no effect recording Sargent polarograph, model XII. The diffusion on the GSH- and GSSO3- wave, served to suppress current of reduced glutathione (denoted as GSH) \vas measthe maximum of the GSSG wave. The polarogram ured manually. All potentials were measured agdinst the 5aturated calomel electrode (S.C.E.). Oxygen was re- resembles that obtained with cystine, except that Ino\-ed from the solutions in the cell with purified nitrogcn the glutathione sulfonate wave is much better and during an experiment an atmosphere of nitrogen \vas defined than the cysteine sulfonate wave which did 11i:iintnined over the solution. The characteristics of the capillary were: ?n = 2.02 mg. not have a well defined diffusion current region sec.-I, t = 3.83 sec., ma/zt'/fi= 2.00 mg.'/r sec.-'/t (open under the Same experimental conditions. A polarocircuit) the height of the mercury column was 50 cm. gram of a mixture a t PH 5 is shown in Fig. 1 R . The pH was measured with a Beckman pH meter, motlrl In solutions of pH lower than 7 sulfite is reduced a t €12. A general purpose blue label glass electrode was usctl for 311 pH measurements. The pH of each solution \ v ~ i s the dropping mercury electrode6 and the waves of GSSG and GSS03 are masked by the sulfite wave. determined a t the temperature specified in Table I . Appropriate volumes of air-free stock solutions of oxi- It is seen from Fig. 1B that the diffusion current dized glutathione and sulfite were added t o a given volume region of reduced glutathione is not affected by the of an air-free buffer solution in an electrolysis cell. The diffusion current of reduced glutathione was measured in the sulfite wave. From the polarographically found concentration mixture after equilibrium was attained. Equilibrium \vas found to be established faster the higher the pH of the mix- of reduced glutathione the apparent over-all equilibture. Thus a t 2.5' a t PH 5 equilibrium was established rium constants of the reactions
after about six hours while a t PH higher than 8 the final state was attained within a few minutes. Diffusion current constants of reduced glutathione were determined at -0.3 and -0.2 volt in solutions of the same compositiori as the equilibrium mixture but in the absence of the disulfide and the sulfonate. The values obtained were used for the calculatioti of the GSH concentration in the equilibrium mixtures. In contrast to cysteine and thioglycolic acid1 thc diffusion current of reduced glutathione a t constant ionic strength ( p = 1) was found to be practically independent of the p H within a pH range of 12.2 to 5 . 2 . At p H 5.2 the GSH concentration should not be higher than 8 X lo-' .lI since at larger concentrations reduced glutathione gives a poorly defined anodic \vave which is preceded by a maximum. Cysteine gives under no conditions a true diffusion current a t pH 3.' The diffusion current of GSH was found to decrease slightly with increasing sulfite concentratioi~. For example a glutathione solution which \\as .I1 in GSH ;it pH 12.15 ( p = 1) gave diffusion currents of 2.63 and 2.53 pa. a t sulfite concentrations of 0.U2 and (J.10 -11,respectively. The temperature coemcient of the diffusion current of GSl1 was determined to be about 2 . 4 5 per degree between 12 and 25' and therefore of the same order of magnitude as that of cysteine and thioglycolic acid. r--T--l
0
1
-0.4
I
-0.8
I
I
-1.2
E . ., v. us. S.C.E.
I
I
-1.6
IbI
I I
-2.0
Fig. 1.-Polarograms at 25' of equilibrium mixtures which were originally: (A) 2 X 10-8 M in GSSG and 0.5 M in Na2SOa in an ammonia buffer (0.1 M NHz, 0.1 M NHICl, 0.9 M K C l , 2.5 X lo-' M thymol, PH 9.23); (B) lo-* M i n GSSG, 0.025 M in NaZSOt in a phosphate-citrate buffer (0.11 M Na2HPO4, 0.045 M citric acid, 0.60 M KCl, PH 5.22). JOURNAL, 61, 825 (1939). (4) J . J Lingane and I. 11.Koithoff. THIS
GSSG
+ SO:'
(SOzH-)
GS-(GSH)
+ GSS0:-
(1)
were calculated for a given mixture. Experimental results and calculated values of the apparent constant K a p p in mixtures of ionic strength one and of various PH and temperatures are listed in Table I. Experiments in solutions of pH markedly higher than 12 could not be carried out with GSSG since the peptide rapidly decomposes in strongly alkaline medium to form reduced glutathione. No experiments were carried out a t PH smaller than 5 because of losses by escape of sulfur dioxide. The values of the apparent equilibrium constant in Table I have been calculated by introducing total concentrations for GSSG, GSSOS-, reduced glutathione and sulfite in equation 1A Kapn
=
[Zred. glutathione] [ZGSSOI-] [ZGSSG][Zsulfite]
(14
where red. glutathione and sulfite are present as GS- and SOy- ions a t pH greater than 8 and as GSH and SOaH- ions a t PH 5 . The dissociation of the sulfhydryl group in reduced glutathione and of the amino group in reduced and oxidized glutathione and in glutathione sulfonate and of sulfite (to bisulfite) vary between pH 5 and 11; therefore the apparent over-all constant Kappvaries with 9H. The apparent acid-base contants of reduced and oxidized glutathione have been reported recently.6 The individual acid-base constants of reduced glutathione were estimated from these dissociation constants6 using a simplified method recently de~cribed.~ The variation of the acid-base constants of reduced and oxidized glutathione and of the sulfonate with temperature was estimated from the apparent heat of ionization AH' of the sulfhydryl and amino group in the same way as for cysteine.' (5) I. M. Kolthoff and J. J, Lingane. "Polarography," Interscience Publishers, New York, N . Y.,1952, p. 559. (6) N. C. Li, 0. Gawron and G. Bascuas, THIS JOURNAL. 76, 225 (1954). (7) W. Stricks and I. M. Kolthoff. ibid , 75. 5673 (1953)
CONSTANTS OF REACTION BETWEEN SULFITE AND OXIDIZEDGLUTATHIONE 2059
April 20, 1055
TABLE I APPARENT EQUILIBRIUM CONSTANT (EQUATION 1) OF THE OXIDIZEDGLUTATHIONE-SLZPITE SYSTEM AT 12 AND 25' Ionic strength of all mixtures Y = 1 f 0.1. Initial concn. of GSSG,
Temp., OC.
Initial concn. of sulfite,
M
12 12 12 12 12 12 12
2 x lo-' 10-8 10-8 2 x lo-'
25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25
2 x 10-8 10-8 2 x 10-3
M 0.05 .05 .075 .05 .05 .05 .05
10-8
2 x lo-' lo-'
lo-' 2 x
.05 .05 .10 .10 .05 .05 .02 .05 .05 .10
lo-'
10-8
lo-' 2 x 10-3 10-8
2 x 10lo-' 102 x 10-8 10-8
.10 .02 .05 .05 .05 .05 .05 .002 .001 .002 .0075
lo-'
2 x 10-8 lo-' 7.05 x 10-4 5 x 10-4 5.05 X lo-'
5
x
lo-'
9H
12.41 12.41 11.61 11.55 11.55 10.90 10.90 12.15 12.15 12.12 12.12 11.62 11.62 11.30 11.25 11.26 11.20 11.20 10.20 10.15 10.15 9.24 8.4 8.4 5.22 5.23 5.22 5.20
In the pH range listed in Table I the carboxyl group is in the ionized form and its constant need not be considered. From the dissociation constantsa of reduced glutathione (pKi(NH2+) = 8.75, PK:(SH) = 9.65 a t 25') the following values of the individual constants were obtained. Concentrations of the various species are denoted by letters above the symbols. C
GSH
I
NHa" C
GSH
I
NH:+ d
GS-
I
iYH, + b
--
d GS -
I
NH: + b GSH
--
+
I
NHz
i d of red. glutathione, ra.
PKA
= 9 . 0 5 a t 12 and 25'
(2)
9'44 at 12'
{ 9.05 at 25'
+ H + ; PKf,
M
1.96 1.23 1.32 1.86 1.17 2.19 1.35
9.86 x 6.2 x 6.76 x 9.60 x 6.06 x 1.05 x 6.43 x
10-4 10-4 10-4
3.23 1.91 3.69 2.09 3.09
1.24 x 7.34 x 1.46 X 8.29 x 1.21 x 7.25 X 5.72 X 1.20 x 7.17 x 1.45 X 8.20 x 5.97 x 1.24 X 7.41 X 7.74 x 1.45 X 8.19 x 6.11 x 3.85 x 4.54 x 4.88 x
10-8 10-4 lo-' 10-4 10-8 lo-' IO-' 10-1
1.85 1.46 3.05 1.82 3.59 2.03 1.59 3.27 1.96 2.05 3.99 2.26 1.60 1.01 1.19 1.28
No of expt.
0.020 ,020 .019 ,018 ,019 ,024 ,023
19 20 21 22 23 24 25
10-4
10-4 10-3 10-4
,041 ,041 ,040 ,040 ,038 ,039 ,039 ,037 .037 ,039 ,037 ,046 ,042 .043 ,054 ,077 0.075 2.9 2.1 2.6 2.9
13 12 16 15 18 17 0 8
10-4
10-4 10-4 IO-'
lo-' 10-4 10-4 10-4 10-4 10-4 10-4
7 11 10 6 5 4 3 2 1 26 27 28 29
a1
GSSG
/
\
8.96 at 12'
-
( 7)
GSSG GSSG 9.93 a t 12' I +I + NHiNHz I I + H + ; P K ' j 9.54 a t 25' NH:NHz
ZGSSG= al
+ H+;
KSPP
&ad. glutathione,
+ bl + cI = bl
{SI + + 1
(8)
(9)
[H +I
The following dissociation of the glutathione sulfonate GSS08- must also be considered in the PH range investigated.
(3)
U
GS -
I
NE12
+H+;
Zred. glutathione = a
pK6
=
( 4)
+b +c +
The equilibrium constants with reference to the various charge species of reduced and oxidized glutathione and of glutathione sulfonate can now be calculated. CI
GSSG
I
I
+
where a [ H + ] / K : = b d and b = d a t 25'. Equation 6 was used to calculate the concentration of each of the species of reduced glutathione a t a given PH. The dissociation constants of oxidized glutathione are given in equations 7 and 8
NHrNH,
+so:-
a GS-
JJ NHI I
br GSSOa-
+ I
NHz
; KI=c1
bza
1s0:-1 (11)
bi
GSSG
a
GS-
a, GSS0,-
w.STRICKS, I. M. KOLTHOFFAND R. c. KAPOOR
2060
VOl. 77
ered to be equal to Kv. The average value of K v is 2.6 a t 25". I \ sot-J- NH,+ I + I Knowing K I , the constant K s of equation 10 can NHa+NH2 NH2 ' be calculated from equations 6, 9 and 11 a t a lower (13) pH (8 to 9). I n this calculation the accuracy of K1lx = the values of Ks depends greatly on the value of the concentration a2 of the glutathione sulfonate with the charged amino group. Since a2 is determined as a small difference between two relatively large figures (for example: a2 = 7.74 X (ZIGSSO3-) - 7.13 X (b2) a t pH 9.24, expt. 3) the accuracy of a2 is not comparable to the accuracy of our measurements and values of Ks obtained by this calculation showed rather large variations (1.5 X a t pH 8.4 and 7.2 X lop9 a t pH 9.24). caz It is more accurate and simpler to calculate K s (15) Kv = al[HSOt-] from both KI and K v with the aid of equation 19. By proper combination of the above equations The value of Ks of 1.8 X thus obtained was we obtain the following relations used for the calculation of the equilibrium constants KII and Klv. Taking a value of 12,000 cal. for K I T = K,' - KI (16) AH of the amino group,' K s a t 12' is found to be KB 7.4 X 10-lo. A4ssumingthat AH for the amino groups of GSSG, GSH and GSS0,- is the same and that K ~ ( H ~ siso ~ practically ) constant between 12 and 25",it is found that the heat of reaction of the five equilibrium reactions is the same. A H was estimated from the experimental values of K I a t 12 and 25" and found to be 9500 cal. where K ~ ( H ~ s o is ~ ) the second dissociation constant Since most of the calculations of the constants of sulfurous acid. are quite similar to those reported in detail for the I t is seen that the equilibrium constants involv- cystine-sulfite system' specific examples of these ing the various charge species can be expressed in calculations are not given in the present paper. terms of K I and Ks, of the known dissociation con- From equations 12, 13 and 14 the constants K I I , stants of reduced and oxidized glutathione and of K I I Iand K I Vhave been calculated. The values of the dissociation constant K ~ ( H ~ofs osulfurous ~) acid. these constants, the change in free energy ( A F = The constants KI and K v can be found directly RT In K) and the heat of reaction AH for reactions from experiments with solutions of sufficiently high I, 11, 111, IV and V are listed in Table I1 which, for and sufficiently low PH, respectively. comparison, also gives the data for the correspondI t is seen from Table I that the apparent constant ing reactions of cystine and dithiodiglycolic acid. has practically the same value over the p H range The value of Kv for the cystine-sulfite system which between 11.2 and 12.1. Solutions of this pH range could not be determined experimentally and which contain almost exclusively the species with un- was not given previously' was calculated from equacharged amino groups and K a p p of these solutions tion 19 using the corresponding data for cystine and is therefore practically equal to KI. The two disso- cysteine. ciation constants of sulfurous acid are K ~ ( H ~ s=o ~ ) An inspection of Table I1 shows that the values 0.017 and K ~ ( H ~ S O ,= ) 6.24 X a t 2508 and no of KI, KII, KIII and KIV are all smaller than one corrections need be made for the presence of bisul- while K v is larger than one. Comparing reactions fite a t p H greater than 9. In the calculations of the I11 and I1 it is interesting to note that K ~ IisI four apparent equilibrium constants a t anionic strength to three times larger than KII for both RSSR and of one concentrations have been written instead of GSSG. This is remarkable, considering the fact activities. Activity coefficients of the various that the products on the left side of equations 12 ionic species cannot be calculated with a fair accur- and 13, presenting reactions I1 and I11 are identical. acy a t such a high ionic strength. All equilibrium Expressing the equilibrium constant as the ratio of constants referred to and determined in this paper the rate constants of the forward (left to right) and are apparent constants with the exception of the backward reaction (K = k l / k 2 ) the difference betwo ionization constants of sulfurous acid. tween KII and KIIIis explained by the difference in The average values of KI are found to be 0.039 the rate of the reactions between the various charge species of reduced glutathione and glutathione suland 0.019 a t 25 and 12", respectively. GS The value of K v can be obtained from experiments with solutions of p H 5 since a t this pH the fonate. The rate of the reaction between 1 NH2 amino groups of oxidized and reduced glutathione GSSO3are practically all present in the charged form while is thus about 4 times as large as that the sulfhydryl group is uncharged. Almost all of and I NHz+ the sulfite is present as HSO3- ion a t fiH 5 . The apGSGSSOjparent constant at PH 5 can therefore be considand I , assum(8) Norio Yui, Bull. Inst. Phys. C h e w . Research, 19, 1229 (1940); of the reaction between 1 NH3+ NH2 H. V. Tartar and H.H . Garretson, THISJOURNAL, 63, 808 (1941). bl GSSG
+
d GS-
62 GSSOa-
m
+
N,N'-DIHYDROXYETHYLGLYCINE AND ITS COMPLEXES WITH IRON
April 20, 1955
ing that the rate of the forward reaction is the same in both equilibria. The variation in these reaction rates is related quantitatively to the relation between the constants K s and K k (see equations 16 and 17), the charged amino group of the sulfonate ( K s = 1.8 X lop9)being a stronger acid than that of reduced glutathione (K&= 4.5 X 10-lo). The
GSlarger reaction rate between
I
GSSOaand
I
NHz NH3+ may possibly be explained by assuming that the reaction between these two ions is initiated by an exchange of protons between the amino group of the sulfonate and that of reduced glutathione. Comparing the three disulfides GSSG, RSSR and TSST i t is seen that the constant K I is largest for GSSG and smallest for TSST. Thus in alkaline medium the equilibrium involving TSST lies farthest to the left. Experiments to be reported in a subsequent paper on the alkaline fission of the three disulfides in the presence of silver or mercury salts
2061
indicate that cystine reacts the slowest and dithiodiglycolic acid the fastest. It thus appears that the reactivity of the three disulfide compounds toward sulfite is not related to their fissility in alkaline medium. It is interesting to note that in contrast to KI, the values of Kv for the sulfite-disulfide system in acid medium are of the same order of magnitude for the three disulfides studied. This is accounted for by the numerical values of the constants which determine K v (see equation 19 for GSSG and RSSR and equation 201for TSST). (20)
where K(sH)is the dissociation constant of the sulfhydryl group of thioglycolic acid. The numerical values of the ratios of the constant KI and the dissociation constant of the sulffor glutathione and cystine, hydryl group KI/K(sH)for dithiodiglycolic acid) are of the order of magnitude of lo8 for the amino acids and lo7 for TSST. The larger value (lo8) for GSSG and RSSR TABLE I1 is multiplied by the term involving the dissociation EQUILIBRIUM CONSTANTS,CHANGEIN FREEENERGY A F constants of the NH3+ groups (KiKLIKsK; in A N D HEATOF REACTIONS I, 11, 111, IV AND V FOR GLUTA- equation 19) which is of the order of magnitude of THIONE (GSSG) AND CYSTINE (RSSR) AND OF REACTIONS I 0.1, thus making the values of K v of the three diAND V FOR DITHIODIGLYCOLIC ACID(TSST) sulfides of the same order of magnitude. The nuGSSG RSSR TSST merical values of KI and Kv of the three disulfides 120 25' 120 25' 25' indicate that in alkaline medium equilibrium lies Kr 0.019 0.039 0.0060 0.010 0,00047 quite far to the left while in acid medium the equilibKII .0030 ,0062 ,00031 .00052 rium is more favored to the right side. The heat of KIII .012 .025 ,0095 .016 reaction A H which has been determined only for KIV ,018 .037 .086 .15 glutathione and cystine is found to be of the same 1.3 2.6 2.5 4.1 1.4 Kv order of magnitude for the two acids. AFI, cal. -2200 -1900 -2900 -2700 -4500 Acknowledgment.-This investigation was supAFII cal. -3300 -3000 -4600 -4500 ported by a research grant (C-721,C5) from the AFIII cal. -2500 -2200 -2600 -2500 National Cancer Institute, U. S. Public Health AFxvcal. -2300 -2000 -1400 -1100 Service. AFvcal. +140 $570 4-290 +890 +200 AH, cal.
$9500
+7000
MINNEAPOLIS, MINNESOTA
[CONTRIBUTION FROM T H E SCHOOL O F CHEMISTRY O F T H E UNIVERSITY O F M I N N E S O T A ]
N,N '-Dihydroxyethylglycine and its Complexes with Iron1p2 BY P. E. TOREN AND I. M. KOLTHOFF RECEIVED OCTOBER16, 1954 At an ionic strength of 0.5 the acid dissociation constants K1 and Kfof N,N'-dihydroxyethylglycine, (DHEG), are l o - 9 . w and respectively. Ferric and ferrous iron and their mixtures in the presence of excess DHEG give reversible waves a t the dropping mercury electrode; the standard potential of the system is -0.94 v. os. the saturated calomel electrode. Ge FeGz 20H-. Potentiometric measThe electrode reaction in neutral and alkaline solution is FeG(0H)s urements a t the platinum electrode indicate that the electrode reaction in acid solution is probably FeG(OH)* e e F e + + G20H- between PH 5 and PH 3, and Fe+' e Fi F e + + in more acid solutions. The complex stability constants determined from electrometric data are KIII = loa*' and KII = 105.0 for FeG(0H)I and FeGz, respectively.
+
+
+
+
N,N'-Dihydroxyethylglycine (DHEG) is an amino acid with the structure RzNCH~COOH,where R = CZH~OH.I n common with glycine and other amino acids, i t is likely that the uncharged form of DHEG exists as a dipole ion, RzH+NCH2COO-. DHEG is a white solid which forms colorless aque(1) This work was carried out under the sponsorship of the Reconstruction Finance Corporation, Office of Synthetic Rubber, in connection with the Synthetic Rubber Program of t h e United States Government. (2) From a P h . D . Thesis submitted by P. E. Toren to the Graduate School of the University of Minnesota, December, 1953.
+ *
+
+
ous solutions. Both the acid and its salt are soluble in water. I n the present paper we report on the determination of the dissociation constants of DHEG and the stability constants of its ferric and ferrous complexes. Equations of the electrode reactions involving the ferric-ferrous complexes of DHEG at the dropping mercury and a t a platinum electrode have also been established. Materials.-The DHEG was furnished by the Alrose Chemical Company. It was purified by dissolving the acid in a small amount of hot water and precipitating with ethanol. A brown color which was not completely removed
