A. B. GARRETTAND
2380
JAMES
LEMLEY
Vol. fj4
[CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY, THE OHIO STATE UNIVERSITY
The Solubility Relations of Mercuric Oxide in Aqueous Solutions of Hydrogen Chloride1 B Y A. B. GARRETT AND JAMES LEMLEU
This paper presents data on the solubility of mercuric oxide (yellow) and basic mercuric chloride (2HgOsHgCb) in dilute solutions of hydrochloric acid. The data on the mercuric oxidehydrochloric acid equilibria are similar to those obtained in a study of the mercuric oxide-nitric acid equilibria2 which involve the hydrogen ion effect, but these have the additional interest in that they include the reaction with the chloride ion. Hence, the following equilibria are involved in interpreting the mercuric oxide-hydrochloric acid data in the very dilute range.
+
__
HgO(s) Hz0 Hg(0H)daq.l (1) HgO(s) H+ 1_ Hg(OH)+ (2) HgO(s) 2H+ Hg++ HzO (3) HgO(s) Hf C1J_ Hg(OH)Cl(aq.) (4) HgO(s) 2H+ 2C1HgClz(aq.) HsO (5) HgO(s) 2H' 3C1- J_ HgCLH20 (fj) HgO(s) f 2H' 4- C1HgCle HzO (7) 3
+ + +
+ +
+ +
+
+
+
+ +
Since the compound 2Hg0.HgClz was identified as the solid phase in the higher concentrations of acid, the following additional reartions are to be considered 3HgO(s) + 2HT + ZC1- J_ 2HgO.HgClz(s) $- H20 (81 aHgO.HgClz(s) f H10 3- H7 4- C13Hg(OH)CI (9) 2HgO.HgCL(s) 4H+ 4C1- J_ 3HgCl~(aq.) 2Hz0
+ + + (10) 2HgO*HgCL(5) + 4H' + C1- J_ 3HgC1' t 2H20 (11) 2HgO.HgCL(d + 4H' + 7C13HgCla- + 2HzO (12)
The preparation of this trimercuric oxychloride is described in this paper. The existence of it is reported in early work.* '1) Original manuscript received Aprrl 5, 1940 (2) Garrett and Howell, THISJOURNAL. 61, 1730 (1939). (3) Other equilibria may be indicated by writing such equatlons A i
2HgO
+
+
HgO C13 2 1 - f Ht0
HgOCl'ZHg(0H)Cl
ca) (b)
However. equations (a) and (b) seem very improbable due to the pronounced hydrogen ion effect in this r-ti0n.r (4) See Mellor, "A Comprehensive Treatise on Inorganic end Theoreticd Chemistry," Longmans and Company, London, 1923, Vol. IV, pages 839-844, for a discuasion of the oxychlorides of mercurs' Reference is piven here to several of the early workers who
The constants for the equilibria represented b?. equations (l),(2), and (3) have been evaluated.' Procedure.-The general procedure is identical to that described in earlier work.2 The liydrochloric acid used was Grasselli c. P. acid. The samples from which the data were obtained for 'Table I1 were made by adding the same amount of Baker and Adamson Reagent mercuric oxide (6.000 g.) to each flask t o which was then added 200 ml. of the standard hydrochloric acid solution. This procedure was followed in order that we might attempt to observe the progressive change in the solid phase as well as the change in solubility. The samples from which the data were obTABLE I SOLUBILITY OF YELLOW MERCURICOXIDE IN HYDROCHLORIC ACID Varying amounts of HgO and HCI. Samples of mercuric oxide were washed repeatedly in conductivity water previous to preparation in order to remove fine particles. Moles HCI/ 1000 g. HsO
x x 3.0 x 3.0 x .i.o x 7.0 x 9.0 x 9.0 x 1.0 x 1.1 x 1.3 x 1.7 x rj.0 1.0
1.9 2.0 4.0
x
x x 5.0 x 5.0 x
10-6 w - 4
10-4 10-4 10-1
lo-.' 10-4 10-4 10-3 10-3 10-3
x
2. 4 i a 2.52 :3 . 7 4. 1 4.9 6.5 8.4 7.1 6 ,7
7.2
Moles HCI/ 1000 g. Hz0
4 . 0 x lo-.' 8.0 lo-' 8 2 X lo-? x 4 x lo-' 8 . 6 x lo-' 9 . 0 x 10-5 1 . 0 x 10-2 1.0 x 10-2 1.0 x 10-2
x
1 . 2 x 10-2 3 . 0 x 10-2 3.0 x 10-2 8.0 x 10-2
10-k' 10-3
9.8 11.i 13.0 13.1
10-s lo-' 10-3
25.1 26.2 25.8"
1.2
10-3
29.6 36.6
18 X 2.0 x
lo-' lo-'
28.7 90.1''
2.8
10--1
lo-"
6.0 X
7.0
10-4 X moles HgO (Wow)/ 1000 g. Ha0
1.0 x lo-$ 7 . 5 x 10-3
1 .o x 10-1 I 2 1.4
x x
x
x
10-1 10-1
10-1
-1.0x lo-'
10.4 X moles HgO (yellow)/ 1000 g . HsO
18.7 36
18.1 18.6"
-11.3 19.0
18.5 56 50.3 31
18.9 16.3 17.7 19.8"
22.8'' 3-1" 47" 232'' 114" 534' 14908
5 . 0 x 10-1 11808 * and * indicate approach to equilibrium from the side of saturation and unsaturation, respectively: these symbols are also used in Tables I1 and 111. crystallized 2HgO.HgCla from a hot aolution containing mercuric oxide and mercuric chloride. See Roucher, Ann. chim. pbys., 131 17, 353 (1849); Compf. rend., 19, 773 (1844); Thummel, Arch Phorm., 9S3, 919 (1888); 217, 589 (1889); Schroek, THISJ O V R N A I 29, 332 (1903); and Droit, Conrpf. vend.. 181, H60 (19111.
Oct., 1942
SOLUBILITY
RELATIONSOF MERCURIC OXIDF
IN HYDROCHLORIC
ACID
2381
tained for Table I were made from Baker and Adamson Reagent mercuric oxide (yellow) which was washed ten times with conductivity water before use.
analyzed was 85.5%, the chloride analysis was 9.9% (theoretical, Hg 85.4'3, and C1 10.06%). The data are given in Tables I, I1 and I11 and are shown graphically in Figs. 1 and 2 . All The Basic Mercuric Chloride, 2Hg0.HgC12 values are expressed in moles per 1000 g. of (Trimercuric Oxychloride).-The black solid water. phase in the samples (Table 11)containing 0.1 A; ___ ' ---gs..I.al to 0.28 N hydrochloric acid was identified by mercury and chloride analysis as conforming t o the formula 2HgO.HgClz. Basic mercuric chloride was prepared by adding mercuric oxide to 0.15 iV solutions of hydrochloric acid, in a slight excess of the amount sufficient to convert the mercuric oxide t o the basic salt in accordance with the equation 3HgO(s)
+ 2HC1
2HgO.HgClz(s)
+ HzO
(8)
The mixture was stirred vigorously, during which time its color turned from a yellow to black; this color change usually occurred within the first two hours. The solution was then decanted and the black solid washed repeatedly with conductivity water containing a small amount of hydrochloric acid. The samples used to obtain the water solubility (Table 111) were washed carefully with water before they were made. The mercury analysis of four of these samples
L
,
,
,
Scale A 0.0010.003 0.004 0.006 0.008 0.01 Scale B 0.0001 0.0002 0.0004 0.0006 0.0008 0.001 Moles HC1/1000 g. HzO. Fig. 1.-Curves 1 and 3, HCl as samples were made up: Curve 2, HC1 corrected for HCI added by reaction due to the equation 2Hg0.HgClz(s) H,O = 3Hg0 2HC1
+
+
_----*
,
,
TABLE I1 SOLUBILITY OF YELLOWMERCURICOXIDE IN HYDROCHLORIC ACID 200 ml. of standard HC1 solution added to each saniplc containing 6.000 g. of dried reagent HgO. 10-4
Moles of HCl/ 1000 g. Ha0
0.
oolo~r
.0030U .00500 ,00700 ,0090 ,0120 ,0160 .0200 .030@ .0400
.oao
.lo07 ,1210 ,1412 ,1614 ,2018 ,2425 2832
x
moles of fiH HgO/ (glass 1000 g . of H a 0 electrode)
9.0 20. 8u 3.18 40.i 28.9 30.9 29.1 22.1 15.0 10.0 31.6 82" 237 38i 545 880 1150 1410
5.0 4.9 4.9 4.7 4.9
5.1 5.0 4.8
5.0 5.2 5.1 4.0 4.1 4.2 4.0
3.8 3.9
Character of solid phase Color %Ha
Yellow Yellow Yellow Yellow and black Yellow and black Yellow and black Yellow and black Yellow and black Yellow and black Yellow and black Brown Black Black Black Black Black Black Black
0.05 0.075 .08 01 Moles of HC1/1000 g. HzO Fig 2 - f , Table I1 data corrected for HC1 tiecessary to convert HgO to 2HgO.HgClt; 0,Table 111 data; data corrected for the amount of HCl added due t o the reaction 2HgO.HgClz(\) Hz0 = 3Hg0 2HCl 0 01
+
85.4
85.5 85.6 85.4
Theoretical normality of 200 ml. of HCI to change 6.000 HgO to 2HgO.HgCls is 0.0922 N . Theoretical % ' of Hg in 2HgO.HgClg is 85.4%.
g.
+
The data i n Table I1 show a definite change in the solubility which is accompanied by a change in the solid phase from yellow HgO to black 2HgO.HgClz. This indicates that reactions represented by equations (4), (5), ( 6 ) and (7) are applicable to the data only below the region of the change; above this region and up to m H C l = a new phase is being eqUatio1l 8; above r n ~ c=~0.1 a new equilibrium is estab-
-%. B. GAXRETT AND JAMES LEMLEY
2882
L*ol. (i-l
lished which may be represented by equations mHCl = 0.0922 should convert the 6.00 g. of mercuric oxide t o 2HgO.HgCls. It is to be observed (H), (lo), (ll),and (12). The value of the water solubility of this basic from the data in Table I1 that the solid phase chloride (2HgOaHgC12) is 11.4 X (Table 111). becomes homogeneous a t approximately 0.1 ;LI hydrochloric acid and the inflection of the soluTABLE 111 bility curve appears here. Additional evidence SOLUBILITY or HASIC MERCURIC CHLORIDE %HgO,HgClt is obtained from the data in Table I11 also shown I N WATERAND I N DILUTESOLUTIONS OF HYDROCHLORIC in Fig. 1. In this experiment samples of 2Hg0. 9CID (Calculated as moles of HgO t u compare with data in HgClz were prepared by the method described Table 11.) above, and their water and acid solubility deterhiolality of HCl a t mined. These data were then plotted on the same 10 X equilibrium Moles moles (molality of pH Solid phabe graph (Fig. 1) as is used for the data in Table IT. of HCI/ of HgO/ HCI a t start (glass analysis 1000 p 10W.l $. +,*/sn HgO. elecHg. $1, In order to compare these two sets of data the HI0 HyO (See Eq. 8) trode) c ordinate is moved to the right t o mHCl = 0.0922 I . oo 11.9 00 12.5 iiZHC1 (due to solution of 2HgO.HgClZ) which is . 00 12,:i the point a t which all the HgO should be con. 00 12.3 verted to 2HgOaHgCla and a t which the data . 00 IU.5 from Tables I1 and I11 should be comparable. .oo 11.4 The agreement is excellent. .00 11.6 .00 11.3 The equilibrium constants, K4, Klo, KL, and .oo 11.6 K n , are difficult to evaluate due primarily to the . 00 12.1 uncertainty of the value of aa-. Some informa.00 9.9 tion can be obtained about the relative distribu.00 11.4 tion of the dissolved mercury among the forms .oo 9.X .00 13.0 Hg(OH)Cl, HgC12, HgCl+ and HgC13- above the 00 10.0 transition point if one makes the safe assumption .00 10.5 that the two mercury bearing ions HgCl+ and .00 11.2 HgC13- only appear in appreciable concentra.00 11.9 tions a t higher concentrations of mercury. The .00 11.4 relative distribution of the dissolved mercury Average 11.4 0.00076 between Hg(0H)CI and HgCla above the transi 0.00200 35.8 0.00372 1.1; 8 d . 2 $1.9 t i r m point is given in Table IT'. ,00364 4.9 .00200 21.7 .i
+
?;
I
.00300 ,00500 . CjO500 .00701 ,00801 ,01002 .03003 ,0501
.ow1
.IO08
32,5 48.2 49.8"
62.2 68.4 87.8 238 385 626 760
,00517 ,00821 ,00831 ,0112 ,0126 ,0159 ,0459 ,0757 ,122 .151
1.S 5.1
85.2
-1.7 5.0 4.6
4.6
S4.S
9.8
4.6 4.2 4.2 85
TABLEIT' APPROXIMATE DISTRIBUTION OF DISSOLVED MERCURY ~h THE REACTION OF 2Hg0.HgC12 WITH HC1 Calculated from the equation below, smooth data from graph Y .> b HCI transformed 2HgO.HgCln to HgO.HC1 c e>(pressed as 10 assumin only Moles of added HC1/1000 g. H20
I
O
Solid phase removed from flasks of samples, reported in Table 11, containing HC1 varying .from 0.12 to 0.28 IC'; this material was washed with conductivity water previous to making these samples : theoretical yo Hg in 2HgO.HgCl2, S5.4a/0; theoretical yo C1 in 2HgO.HgCl9, l O . l % . Above wiIlCl = 0.1 Table I1 only one phase (black) is observed. This phase gave the mercury and chlorine analysis in agreement With the formula 2Hg0.HgC12. Further evidence of the identity of this phase is the fact that 200 ml. of
moles HgO/ 1000 g. H i 0
0.001 0015 002 004
18.4 22 0
49 32
25 6
24 1''
01 02
40 3 35 3 70 0 85 8 161
04
808
006 008
s=
HgO.HS1 and HgCh are formed
( S C)
\
h
1 d
1
- 0 75
2 25
The Transition.-No definite statement can be made of the exact value of mHCl a t the
Oct., 1942
ADSORPTION AND ENERGY CHANGES AT CRYSTALLINE SOLIDSURFACES
transition point. The problem yet t o be solved is to determine the equilibrium concentrations of HCI, HgO and 2HgOsHgCl at the triple point.
Summary The solubilities of mercuric oxide and of basic mercuric chloride, 2HgO.HgC12, have been measured in hydrochloric acid solutions. A break oc-
[CONTRIBUTION FROM
2383
curs in the solubility relationships due to the formation of a new solid phase (2HgO-HgCt). This basic mercuric chloride, 2HgO.HgCl2, has been prepared and identified; the water solubility is 11.4 X lo-*. The relative distribution of mercury among several different ion species is indicated. RECEIVED JULY 25, 1942
COLUMBUS, OHIO
DEPARTMENT OF CHEMISTRY, UNIVERSITY
OF CHICAGO]
Adsorption and the Energy Changes’ at Crystalline Solid Surfaces BY G. E. BOYDAND H. K. L I V I N G S T O ~ ~
Theory
Obviously, is defined by
Adsorption is accompanied by a change both in the total and in the free surface energy of a surface. The classical relation of J. W. Gibbs2for interfaces between fluids has been used extensively and certainly must be regarded as valid when properly applied. The question of the applicability of the theorem to the interface between solids and fluids must be considered, and we are fortunate that Gibbs has discussed the problem in detail in the case in which the solid is anisotropic. Thus, if the total surface energy, Eso; surface entropy, Sso; surface densities of chemical species, ri; chemical potentials, pi; and temperature, T , are used, it is possible to define a quantity, {, for the arbitrarily restricted case of a system of two components by the equation3 I = E,, - rs,, - plrl - p2r2 (1) In Gibbs’ words, “The quantity 5 evidently represents the tendency t o contraction in that portion of the surface of the fluid which is in contact with the solid. I t may be called the super6cial tension of ajluid in contact with a solid. Its value may be either positive or negative. “It will be observed for the same solid surface and for the same temperature but for different liquids the values of Ysf (in all cases.to which the definition of this quantity is applicable) will differ from those of { by a constant, viz., the value of Ysofor the solid surface in a vacuum.” (1) Original manuscript received February 4, 1942. ( l a ) Present address: Jackson Laboratory, E. I. du Pont de Nemours and Co.. Deepwater, N. J . (2) J. W. Gibbs, “Collected Works,” Longmans, Green & Co., New York. N. Y . , 1928, p. 314. (3) It will be noticed that we have taken the liberty of modifying part of Gibbs’ original nomenclature in order to bring it into approxi-
.
.
. . .
f = Y.f
- Y,,
=
(2)
-?r
or, the spreading pressure, 7 , is equal to the difference between the free surface energy of the clean solid surface (in a vacuum) and the free surface energy when in equilibrium with a chemically dissimilar fluid component (i. e., gas or liquid). Utilizing Gibbs’ general thermodynamic methods, from equation (l),it is possible to obtain for a system consisting of a crystalline adsorbent and one adsorbate da = SwdT
+ I’ldgl + r2dg2
(3)
If isothermal conditions are maintained, and if the Gibbs plane from which adsorption is reckoned is chosen so that the surface density of adsorbent, rl,is zero, one may write d?r = I’p(’)dpp
or
For the case in which the fluid contiguous to the solid surface is a gas or a vapor dgz = RT d In fp
(5)
where fi is the fugacity, and for vapors a t low pressures RT d In fz = R T d 111 pd (6) is true with sufficient accuracy, so that equation (4)may be written as d?r = R T r & l ) dIn p , (7) where pz is to be taken as the equilibrium pressure of a gas or vapor above the crystal surface upon which adsorption has occurred. lo , (4). Extensive . cdlculations .
- . - - - -. .-
eatabliah the correcturas uf
. .
..
Oqii