are used. The inactive meso form has a dispersion ratio of 2.06 and Xo of approximately 3200 8. In Table I the available data are summarized. Previously published data were used in calculating the dispersion ratio a t 4705 and 5880 8.for some of the compounds. TABLE I DISPERSIONRATIOOF METAL-DIAMINECOMPLEXES Complex
(1)
(2)
Reference
P(706/L16880
Complexes containing more than two molecules of an optically active diamine
[ R h ( l - ~ h x n )Cls ~] [ R h ( E ~ p t d i n )Cls ~] [Rh(d-~tn)alI3 [Pt(l-pn) 4Br4 [Pt-(d-pn)zlClz [Zn(l-pn)?d CIS [Cd(Z-pn)z-a](NOa)z
1371
COMPLEX IONSOF ZINC AND HYDROXYLAMINE
March 5 , 1935
1.75 (6) 1.74 (7) 1.75 (12) 1 . 7 4 This work 1.74 (11) 1 . 7 4 This work (av. of 3 results) 1.75 This work (av. of 3 results)
Complexes containing less than two molecules of an optically active diamine
[Pt(l-phe~ien)(en)]I~ 1.80 [Pt(Htapn) C141C1 1.88 [Pt(l-pn)(NH&]C12 1.83 [ C d ( l - p n ) ~ . ~ , ] ( X O ~ ) ~1 . 8 8 [Zn(l-pn)l.851 C ~ Z 1.85 [Ag(Z-pn)l.~1 (KOd 1.85
(13) (14) This work This work This work This work
(13) H. Reihlen, G. Seipil and E. Weinbrenner, A n n . , 620, 256 (1935). (14) F. G . Manu, J . Chem. Soc., 1221 (1927); 890 (1928).
[CONTRIBCTION FROM
THE
(3) Complexes with inactive diamines [Rh(en)a1 Is 1.44 (15) [Irfen),] Br: 1.56 (15) [Pt(en)d Clr 1.68 (16) [Rh(m-ptn)s] Ia 2.08 (12) [Pt(rn-stien)(buten)]Cl2 1.66 (17)
Conclusions From the results reported here the following conclusions seem justified : 1. The source of optical activity in metal complexes with optically active diamines is different than for the resolved metal complexes with inactive diamines. 2. For the complexes with optically active diamines : A. The dispersive power seems independent of the nature of the metal atom even when the metal atom itself may become a center of asymmetry. B. The dispersive power seems independent of the nature of the optically active diamine. C. The dispersive power appears dependent on the formation function e. The ratio a4705/a5880 is 1.74-1.75 when e > 2 and 1.85when il. < 2. Finally because of the constant dispersion ratio for the optically active amines and for the metal complexes with these amines, it seems fair to presume that the optical activity of the complex arises only from the ligands, and not from any induced asymmetry around the central atom. (15) F. M . Jaeger, R8G. t m o . chim., 38, 11 (1919). (18) J. P. Mathieu, BUN. soc. chim., 6, 1258 (1939). (17) W. H. M ~ l l sand T. H. Quibell, J . Chcm. Soc.. 839 (1935).
MINNEAPOLIS, MINNESOTA
DEPARTMENT O F CHEMISTRY, STATE COLLEGE
OF
WASHINGTON']
Complex Ions of Zinc and Hydroxylamine BY C. J. NYMAN RECEIVEDJUNE 7, 1954 The complex ions formed between zinc ion and hydroxylamine have been investigated by the polarographic method, and two species, [Zn(NH20H)]++ and [Zn(NH20H)z] ++, were found t o exist in aqueous solution. I n solutions of ionic strength 1.0, the formation constants of the ions were found to be 2.5 f 1 and 10.2 f 2, respectively. No evidence was found t o indicate the presence of any higher species of complex ions up t o 1 M hydroxylamine concentration.
Introduction Early reports in the literature by Crismer2 and by Antonoff3 indicated the preparation of the solid coordination compound ZnC12.2NHzOH. GoldSchmidt and Syngros4 reported the preparation of ZnCO3-2NHZOH. These authors also reported the presence of the ion [Zn(NH20H)z]+t"in aqueous solutions of zinc salts and hydroxylamine, but made no attempt to determine stability constants. Experimental The procedures and equipment employed in this investigation were the same as those employed in another study.6 (1) This project was carried out under support of the O 5 c e of Ordnance Research, U. S. Army, Project N o . DA-04-200-ORD-65, and the State College of Washington Committee on Research. (2) L. Crismer, Bull. soc. chim., 3, 114 (1890). (3) G . Antonofi, J . Russ. Phys. Chem Soc., 37, 476 (190.5). (4) H. Goldschmidt and K. L. Syngros, 2. anoig. C h c i q . , 6, 1211 (1894). ( 5 ) C. J. Nymnn, T E I EJOUBNAL, 76, 3575 (1953).
All chemicals were of reagent grade and were used without further purification with the exception of hydroxylamine hydrochloride. This salt was purified by recrystallization from 95% ethanol. Stock solutions of 1.920 M hydroxylamine hydrochloride, 4.02 X 10-2 M zinc chloride, 1.9 11.1 potassium hydroxide and 4.0 M potassium chloride were prepared. All solutions used in the polarographic runs had a zinc ion concentration of 4 X 10-4 mole per liter, and 1.0 M potassium chloride in addition to varying amounts of hydroxylamine. The solutions were prepared from the stock solutions by converting hydroxylamine hydrochloride t o free hydroxylamine just prior t o use by the addition of potassium hydroxide in such quantity that there was still a small amount of hydroxylammonium ion present. Sufficient potassium chloride was added so that on the final dilution t o volume, the ionic strength of the solution was 1.0. The p H of the diluted solution was measured with a Beckman Model G pH meter, and the free hydroxylamine was calculated by consideration of the equilibrium NHzOH.H+ NHzOH HC with equilibrium constant 1.04 X 10-6.5 It was assumed
+
(6) H. Hagisawa, Bull. Znst. Phys. Chem. Research (Tokyo), SO, 251 (1941).
C. J. NYMAN
1372
Vol. 77
that this value for the equilibrium constant was valid at ionic strength 1.0, and that the activity coefficients were unity. Data presented by Hagisawaa and Winkelblech' indicated that the value of the dissociation constant did not change markedly with increase in ionic strength of the solutions. At the pH of these solutions, practically all of the hydroxylamine is present as the free amine. In view of the small variation of half-wave potential with hydroxylamine concentration, and the possible error involved therein, this estimation of hydroxylamine concentration is acceptable.
small values of the formation constants obtained, it was thought necessary to consider the hydrolysis of zinc ion. The formation constant of [ZnOH]+ has been determined by Kolthoff and Kameda.s Recently, Fulton and Swinehart lo have reported equilibrium constants from which the formation constants of the other hydroxide complexes of zinc may be calculated. These are summarized below.
Data and Discussion
Zn++ OH[ZnOH] +; B1 = 2.45 X 10' Zn++ 2 O H - z [Zn(OH)z]; Bz = 5.72 X 10"
In Table I are recorded half-wave potentials and the diffusion currents of zinc ion with various concentrations of hydroxylamine and the fiH of the various solutions.
+ +
Zn++ f 3 O H - z [Zn(OH),]-; Bs = 1.71 X 10" Zn++ f 4 O H - z [Z~IOH)~]--; B4 = 3.14 X 10l6
DeFord and Hume related the shift in half-wave
TABLE I potential to the concentration of the complexing VARIATIONOF E ~ , FOR z ZINC IONWITH INCREASING CON- agent by the equation CENTRATION OF HYDROXYLAMINE IN 1.0 M POTASSIUM -!?J(Cx)j = CHLORIDE SOLUTION AT 25' Hydroxylamine, moles/l.
@-amperes
7.10 7.51 7.52 7.57 7.57 7.58 7.01
2.10 2.10 2.11 2.11 2.10 2.05 1.98 1.94
..
0.00
.093 .187 .280 .373 .466 .653 .913
id,
pH
-
El/, v. ws. S.C.E.
1.0127 1.0164 I. 0207 1.0248 I . 0282 1.0317 1.0368 1.043
The method of DeFord and Hume8 was used to determine the species of complex ions in solution and their formation constants. In view of the
i
antilog (0.435
nF [(EOl/p).-
+ log 2)
(1)
I n this expression, Kj is the formation constant for the reaction Zn + +
4-~ N H z O H
[Zn(NH20H)jI
++
C, is the concentration of hydroxylamine in moles s (I3/Jcare the half-wave poper liter; ( E o ~ / Jand tentials of the simple and complex zinc ions, respectively; and I, and I, are the diffusion current constants of the simple and complex zinc ions, respectively. In order to take into account the hydrolysis of zinc ion, one must revise equation (1) to the following nF
(
CRj(C,)f = antilog 0.435
12
(Ev,)J
log & / I c )
rT[(Eo1/2).-
+
- [&(OH) 4- &(OH)* 4- &(OHIS + BdOH)'] (2)
10
8 4 v
K 6
4
2
0.2 Fig. 1.-l'alues
0.4 0.6 0.8 IiHzOH (m./l.). of Fo(X), F , ( X ) and F * ( X ) for zinc ion in hydroxylamine solution Z p h y s i k . Chem., 86, 574 (1901). and D. N. Hume, THISJ O U R N A L , 73,5321 (1951).
( 7 ) K. Winkelblech,
(8) D.
L). DeFord
where B1 through Bd have the values given above and (OH) represents the concentration of hydroxide ion in the solutions. Setting the right-hand side of equation 2 equal to Fo(X), the subsequent operations as outlined by DeFord and Hume can then be carried out. The quantity F,(X) is defined as equal to [Fo(X) ll/Cx, and Fj(X> = IF(j-d(X) - K(.i-dI/G. If Fj(X) plotted against C, and extrapolated to C, = 0, the value of Fj(X) at the intercept is Kj . It was found that the two species [ZnOH]+ and [Zn(OH)z] are present in concentration sufficiently great to make a slight correction to the value of Fo(X),while the concentration of [Zn(OH)a]- and [ZII(OH)~]-- is so small that their effect is negligible. Plots of Fj(X) for the zinc-hydroxylamine complexes in a supporting electrolyte of potassium chloride are shown in Fig. 1. It is apparent from these plots that the two ions which exist in solution under the conditions of this investigation are [Zn(N'HzOH)]++and [Zn(NH20H)2]++with formation constants Kl = 2.5 + 1 and Kz = 10.2 + 2. The fact that the plot of F2(X) is a straight line with a slope of zero indicates that the species (9) I. M. Kolthoff and T. Kameda, ibid., SS, 835 (1931). (10) J. W. Fulton and D. F. Swinehart, ibid., 76, 864 (1954).
1373
March 5, 1955
NOTES
[Zn(NH20H)2]++is the most complex which exists in the solutions under investigation here. The finding that this ion is capable of existence is in agreement with the reports of the formulas for
the solid coardination compounds and with the previous report of the existence of this ion in aqueoussolution. PULLMAN, WASHINGTON
NOTES Determination of n- or p T y p e Conductivity by the Effect Produced by Hydrogen Adsorption on Electrical Conductivity B Y LOUISF. HECKELSBERG, GRANTc. BAILEYAND ALFRED
exhibit this behavior are as follows: nickel oxide, chromium sesquioxide, stannous sulfide and chromiaalumina were classified as p-type semiconductors while titanium dioxide, zinc oxide, zinc sulfide, calcium oxide and tungsten sulfide were CLARK classified as n-type semi-conductors. The change RECEIVED JULY 6, 1954 produced by hydrogen on conductivity has been Introduction.-In the investigation of the elec- reported previously for stannous sulfide, chromiatrical and catalytic properties of semi-conductors alumina8 and for zinc oxide.‘ These results agree with the classification of these materials by Hall the determination of the mode of conductionwhether it is n-type or p-type-is of interest. At effect as reported in the literature. the present time the mode of conduction is deterIn the case of insulators such as silica this effect mined quantitatively by measuring the Hall effect was not observed. The magnitude of change observed was found to or qualitatively by the thermoelectric effect. As a result of observations made during an investiga- vary from a ten-thousand-fold change for titanium tion of the electrical and catalytic properties of dioxide to a threefold change for zinc sulfide. powdered metal oxides and sulfides, we have found The effect produced by hydrogen was found to another method which can be used for the qualita- be reversible. After the hydrogen contacted the tive determination of the mode of conduction of semi-conductor the material’s original conductivity could be restored by flushing the hydrogen out with these materials. This method involves the effect produced by ad- nitrogen. The change produced on the electrical sorbed hydrogen on the electrical conductivity of conductivity by the nitrogen flushing was much these materials. We have observed in the case of slower than the change produced when hydrogen the p-type semi-conductor chromium sesquioxide contacted the semi-conductor. that its conductivity is drastically reduced as rapDiscussion idly as hydrogen can come in contact with it. The Pretreatment of n-type semi-conductors, such as converse effect was observed for the n-type semiconductor zinc oxide where the conductivity was zinc oxide, with hydrogen results in an increase of drastically increased when hydrogen came into electrical conductivity caused by an increase in contact with it. As a result of these observations lattice defect concentration. These defects are it was decided to check several additional semicon- formed by the oxygen or sulfur combining chemiductors to see if this effect was of a general nature. cally with the hydrogen leaving a cation-excess maExperimental Details.-As an example of this method we terial. The increase observed by reduction with shall describe the procedure followed for nickel oxide. hydrogen is slow compared to the rate of increase The sample of nickel oxide was pretreated with oxygen at observed when hydrogen is adsorbed on a semi500’ for 12 hours in an apparatus previously described by the conductor in an inert atmosphere. authors? With prepurified nitrogen flowing over the nickel A plausible explanation of the mechanism inoxide its conductivity was measured as a function of temperature. At 300’ hydrogen was introduced over the nickel volved in reducing the current flow in p-type semioxide. This resulted in a rapid decrease in the conductiv- conductors is as follows. When the hydrogen ity of the nickel oxide of the order of about lo-*. After comes into contact with a p-type semi-conductor it approximately one minute the conductivity began to in- dissociates and ionizes. The protons are adsorbed crease as the nickel oxide was being reduced. This may mean that the nickel oxide on contact with hydrogen under- on the surface while the electrons enter the lattice went a transition from p-type to n-type and ultimately to a and combine with positive holes. Filling the positive metallic conductor. In the case of chromium sesquioxide holes-which are considered to be the current carwhere hydrogen does not materially reduce the oxide at riers in p-type semi-conductors-reduces the cur500’ the conductivity decreases and remains extremely low. rent flow. Results.-The results for semi-conductors which The mechanism involved in the increase in curwe have tested or were tested by others and that rent flow observed when hydrogen comes into con(1) Heckelsherg, Clark and Bailey, “A Correlation between Lattice (2) J. s. Anderson and M. C. Morton, Proc. Royat SOC.London, l84A Defects and Catalytic Activity of Zinc Oxide” presented before a Symposium on Mechanisms of Homogeneous and Heterogeneous Hydrocarbon Reactions at the Kansas City Meeting of the American Chemical Society, March 29 to April 1. 1954.
83 (1946). (3) P. B. Weisz. C. D. Prater and K. D. Rittenhouse, J . Chcm. Pkyr., 42, 2236 (lQ53). (4) P. Stockmann, Z. Physik, la7, 563 (1950).
