THECRYSTAL STRUCTURES OF BISMUTH HALIDECOMPLEX SALTS
3117
The Crystal Structures of Bismuth Halide Complex Salts. 111. Tris(dimethylammonium) Hexabromobismuthate (111) ,
by W. Gant McPhersonl and Edward A. Meyers Department of Chemistry, Texas A & M University, College Station, Texas 77843 (Receiued November 99, 1967)
[(CH$zNHz]rBiBreis rhomboheiral, R3, with z = 12 in the unit cell indexed on a hexagonal basis; a = b = 29.25 (6) and c = 8.45 (2) A. The structure was determined from the intensities of 1160 independent (ohkl) reflections, 1 = 0-4, collected with a Weissenberg camera, with (Ni-filtered) Cu K a radiation (A 1.5418 A) and multiple-film packs. The crystal was treated as a cylinder, p R = 3.12 for absorption corrections. Full-matrix least-squares refinement of the parameters gave R1 = ZIF, - Fol/ZIFo\ = 0.11 with isotropic temperature factors, and RI = 0.08 with anisotropic temperature factors for Bi and Br. I n the least-squares refinement, 80 reflections with lF,l 2 325 were given zero weight and the remaining 1080 reflections were given unit weight. Two crystallographically independent Bi(III)Br6 octahedra, (A) and (B), are present in the structure. For (A), Bi(A)-Br(A) = 2.849(7) & . and Br(A)-Bi(A)-Br(A) = 92.5 (3)'. For (B), Bi(B)Br(l)(B) = 2.822 (9) A, Bi(B)-Br(2)(B) = 2.852 (8) A, Bi(B)-Br(3)(B) = 2.839 (8) A, Br(l)(B)-Bi(B)Br(2)(B) = 91.9 (3)', Br(l)(B)-Bi(B)-Br(3)(B) = 87.2 (2)', and Br(2)(B)-Bi(B)-Br(3)iB) = 90.1 (2)'. Thus the Bi-Br bond distances do not appear to differ significantly from their average, 2.840 A, but the angular deviations from 90.0' indicate that the octahedra are not regular. The total Pauling bond order is 2.7 for the Bi-Br bonds in an octahedron. (No satisfactory corrections were made for the effects of thermal motion on bond distances, so that the values cited may be too small and less reliable than indicated.) Two crystallographically independent cations, A and B, are present in the structure. The mean C-N bond distance is 1.56 A, the mean C-N-C bond angl: is 118', with no significant deviations from these values. The van der Waals contacts are greater than 3.6 A, except for Br(3) (B)-C(2) (B) = 3.47 (9) 1. There are short N Br contacts. For N(A), N(A)-Br(A) = 3.44 (7) wand N(A)-Br(2)(B) = 3.48 (7) A, and jhese contacts couple the two types of Bi(II1)Bre octahedra together. For N(B), N(B)-Br(l)(B) = 3.49 (6) A and N(B)-Br(2)(B) = 3.44 (6) A, and these contacts are exclusively to the Bi(I1I)Bre octahedron, (B).
Introduction In two earlier studies2rawe have reported the structures of organic ammonium salts of Bi(III)Br4, Bi(II1)14,and Bi(III)Br5. Several regularities were observed in these structures. In each case, the Bi atom was surrounded by a distorted octahedron of halogen nearest neighbors and linked into infinite chains via halogen bridges. Moreover, the total Bi-halogen Pauling bond order4 was close to 3. I n the present study, the structure of [(CH3)2NH2]aBiBr6 was determined in order to obtain structural parameters for the anion units which were expected to be Bi(III)Bre octahedra. Experimental Section [ (CH&NHz]aBiBre was prepared by Osborne.s Single crystals were grown by recrystallization from a 1 :1 water-propanol mixture, approximately 1 M in HBr. A nearly cylindrical needle-shaped crystal 0.2 mm in diameter was selected and mounted in a 0.2-mm thinwalled Lindemann glass capillary. Precession photographs with (Zr-filtered) Mo K a radiation (A 0.7107 A) were obtained and the unit cell was found to be rhombohedral. When indexed on the basis of a hexagonal cell with the c axis along the needle direction, a = b =
29.25 (6) 8, C = 8.45 (2) 8, dmeasd = 2.5 (1) g/CC, = 2.63 (1) g/cc for x = 12. The crystal was transferred to the Weissenberg camera and (hkl) data were collected for 1 = 0-4 with (Ni-filtered) Cu K a radiation (A 1.5418 A) and a multiple-film pack. The value pR = 3.12 was used for the absorption corrections for the cylindrical sample, and Lorentz and polarization corrections were also applied to the approximately 1160 reflections measured with a Welch Densichron, Model 10. Only reflections IC 1 = 3N were observed, which for which -h indicated a rhombohedral space group. Refinement was begun in space group R3mS6 Three Bi atoms were generated from Bi(A) placed in special dcslod
+ +
(1) Submitted to the Graduate College of Texas A & M University in partial fulfillment of the requirements for the degree of Doctor of Philosophy, Jan 1967. (2) B. K.Robertson, W. G. Mopherson, and E. A. Meyers, J . Phys. Chem., 71, 3631 (1967). (3) W.G. MoPherson and E. A. Meyers, ibid., 72, 632 (1968). (4) L. Pauling, J . Amer. Chem. Soc., 69, 642 (1947). (6) J. F. Osborne, Master's Thesis, Texas A & M University, College Station, Texas, 1960. (6) "International Tables for X-Ray Crystallography," Vol. I, Kynoch Press, Birmingham, England, 1962,pp 272-273, 252-253. volume 72, Number 0 September 1068
3118
W. CANTMCPHERSON AND EDWARD A. MEYERS
Table I : Atomic Coordinates of [ ( C H ~ ) Z N H Z ] ~Obtained B ~ B ~ E in the Isotropic Refinement (Standard Deviations Are Given in Parentheses and Apply to the Last Digit of a Number)
kl
X
1!
z
d
0 0 0 0
0 0 0 0
0 0 0 0
3.50 (10) 3.31 (11) 4.07 (8) 3.87 (8)
a
b C
B,
a
'/a
b
'/a '/a
'/a
'/e
C
1/a
'/s
d
'/a
'/a
'/a
'/a
3.00 (5) 2.80 (5) 3.55 (5) 3.35 (5)
a
0.0852 (2) 0.0852 (2) 0.0852 (2) 0.0852 (2)
0.0766 (2) 0.0766 (2) 0.0763 (2) 0.0763 (2)
0.1860 (9) 0.1860 (9) 0.1868 (7) 0.1868 (7)
4.36 (13) 4.55 (13) 5.01 (10) 5.21 (10)
0.1487 (3) 0.1486 (3) 0.1489 (2) 0.1488 (2)
0.2525 (3) 0.2525 (3) 0.2524 (2) 0.2524 (2)
0.5496 (10) 0.5496 (10) 0.5481 (8) 0.5481 (8)
5.14 (15) 5.33 (15) 5.77 (11) 5.98 (11)
0.1433 (2) 0.1433 (2) 0.1434 (2) 0.1435 (2)
0.2642 (2) 0.2642 (2) 0.2644 (2) 0.2643 (2)
0.0697 (9) 0.0698 (9) 0 0686 (7) 0.0687 (7)
4.47 (13) 4.65 (13) 5.21 (10) 5.39 (10)
0.0581 (2) 0.0581 (2) 0.0579 (2) 0.0579 (2)
0.3016 (2) 0.3016 (2) 0.3015 (2) 0.3015 (2)
0.3620 (9) 0.3620 (9) 0.3615 (6) 0.3615 (7)
4.17 (12) 4.35 (12) 4.79 (9) 4.97 (10)
0.020(3) 0.020 (3) 0.021 (2) 0.021 (2)
0.139 (3) 0.139 (3) 0.138 (2) 0.138 (2)
0.413 (11) 0.413 (11) 0.413 (8) 0.412 (8)
6 . 7 (17) 7 . 0 (18) 7 . 3 (13) 7.7 (13)
0.037 (3) 0.037 (3) 0.036 (2) 0.036 (2)
0.160 (3) 0.160 (3) 0.159 (2) 0.158 (2)
0.263 (9) 0.262 (9) 0.250 (7) 0.248 (7)
8 . 1 (17) 8.6 (17) 9.2 (13) 9.4(13)
0,000 (3) 0.000 (3) 0,001 (2) 0.001 (2)
0.167 (3) 0.167 (3) 0.163 (2) 0.163 (2)
0.149 (11) 0.149 (11) 0.138 (7) 0.138 (7)
8.1(21) 8 . 6 (22) 7.4 (13) 7.7(13)
0.434 (3) 0.434 (3) 0.430 (2) 0.430 (2)
0.125 (3) 0.125 (3) 0.123 (2) 0.123 (2)
0.105 (9) 0.105 (9) 0.104 (7) 0.105 (7)
6 . 3 (16) 6 . 7 (17) 6 . 9 (12) 7 . 2 (12)
d
0.373 (2) 0.373 (2) 0.375 (1) 0.375 (1)
0.118 (2) 0.118 (2) 0.117 (1) 0.117 (1)
0.118 (7) 0.119 (7) 0.102 (5) 0.103 (5)
6.1(13) 6.4 (13) 6 . 1 (9) 6 . 4 (9)
a
0.328 (3)
0.328 (3) 0,332 (3) 0.332 (3)
0.060 (3) 0.060 (3) 0.062 (3) 0.063 (3)
0.174 (10) 0.174 (11) 0.176 (8) 0.176 (8)
8 . 4 (21)
b
b C
d a
b C
d a
b C
d a
b C
d a
b C
d a
b C
d a
b C
d
a b C
d a
b C
C
d
'/a
'
I
Unit weights, all data, corrected for ad. a Unit weighte, all data, uncorrected for anomalous dispersion (ad). Unit weights, selected data (IFol 5 325), corrected for ad. selected data (IF,]5 325), uncorrected for ad.
position a (0,0, 0), and nine Bi atoms from Bi(B), in a special position directly related to e, (l/2, 0, 0), (0, l/2, 0), and l/2, 0). A Fourier synthesis was The Journal of P h y a k l Chemistry
8.9 (22) 10.0 (18) 10.6 (18) Unit weights,
constructed' from (hlcO)data with the Bi atoms included in the calculations.s Only four Br peaks appeared in the vicinity of Bi(B). The space group R%m was
3119
THECRYSTAL STRUCTURES OF BISMUTH HALIDECOMPLEX SALTS Table I1 : Bond Distances and Angles for ((CH&NHt)aBiBrs (Standard Deviations Are Given in Parentheses and Apply to the Last Digit of a Number) Table I
Table 1--
From b rows
From d rows
2.849 (7) 2.822 (9) 2.852 (8) 2,839 (8) 1.40 (9) 1.54 (10) 1.68(8) 1.61 (9)
2.850 (6) 2.818 (7) 2,858 (7) 2.844 (8) 1.48 (7) 1.46(7) 1.54(6) 1.59 (7)
From a rows
-Bond distance, A
92.5 (3) 91.9 (3) 87.7 (2) 90.1 (2) 121 (7) 115 (5)
discarded, and R P was selected for all further calculations. The Bi positions were not affected by the change in the choice of space group. The atoms Bi(A), Bi(B), and Br(A), the bromine atom close to Bi(A), were included in the electron density calculation with (hkO) data, and it was apparent that two of the four peaks around Bi(B) were exceptionally large. This indicated that each of the larger peaks was made up of the contribution of two Br atoms superposed along [Ool]. The x coordinate of each Br atom was obtained with the assumptions that Bi-Br = 2.8 A and that the Bi(III)Bre group was a regular octahedron. The Bi and Br coordinates were refined by least squareslgwith unit weights, isotropic temperature factors, and no correction for anomalous dispersion, RI = Z I F , - F,I / Z I F , 1 = 0.14 and Rz = [ Z ( F o- FO)z/ZFO]”* = 0.16. I n order to locate the positions of the light atoms (C and N) of the two crystallographically distinct dimethylammonium groups required, a three-dimensional difference Fourier synthesis was calculated, with the contributions of the heavy atoms (Bi and Br) subtracted. Additional least-squares refinement was carried out, with C and N coordinates permitted to vary. Improvement was noted, R1 = 0.120 and Rz = 0.128. A similar set of calculations were made, with anomalousdispersion corrections included,10R1 = 0.120 and RZ = 0.129. The results of these refinements are given in Table I, rows a and b, and the values of some relevant bond angles and distances calculatedll from the b rows are given in Table 11. From the calculations, it may be seen that there were no significant differencesbetween the results obtained with or without the inclusion of anomalousdispersion corrections for the least-squares refinement of all of the data observed, with isotropic temperature factors and unit weights.
92 3 (2) 91.8(2) 88.0 (1) 90.1 (1) 124 (5) 111 (4) I
---
2.851 (5) 2.817 (6) 2.867 (6) 2,843 (7) 1.47(5) 1.50(5) 1.55(4) 1.59 (5)
2.851 (5) 2.818 (6) 2.867 (6) 2.843 (7) 1.47 ( 5 ) 1.49 (5) 1.56(4) 1.58(5)
--
-Bond angles, deg-
Br (A)-Bi(A)-Br(A) Br(l)(B)-Bi(B)-Br(2)(B) Br (1) (B)-Bi( B)-Br (3)(B) Br(2)(B)-Bi(B)-Br(3)(B) C (11(A)-N (A)-C (2)(A) C(1)(B)-N(B)-C (2103)
From b rows
92.3 (2) 91.9 (2) 88.1(2) 90.1(1) 124 (3) 110 (3)
92.3 (2) 91.9 (2) 87.9 (1) 90.1 (1) 124 (3) 109 (3)
The ratio of scale factors for corresponding zones in rows a and b varied from 1.035 to 1.045. During the estimation of intensities, it had been observed that the more intense reflections, IFo[ 2 325, were difficult to read, even on the lightest films. Also, the reflection (113) was found to be partially in the shadow of the beam stop support. This reflection was discarded and the least-squares refinement of the atomic parameter and zonal scale factors for the 1080 reflections that remained was carried out with isotropic temperature factors and unit weights. The results are given in Table I in the c rows for which no corrections were made for anomalous dispersion and in the d rows for which corrections were made for anomalous dispersion. R1 = 0.108 and RZ = 0.119 and R1 = 0.109 and Rz = 0.120 for rows c and d, respectively. The ratio of corresponding zonal scale factors between rows c and d varied from 1.035 to 1.038. The scale factors were then fixed at the values obtained in the isotropic refinements in c and d rows of Table I and the anisotropic refinement of Bi and Br was carried out to give the results in rows a and b in Table 111 and in the last two columns of Table 11; R1 = 0.079, Rz = 0.090, R1 = (7) W. G. Sly, D. P. Shoemaker, and J. H. Van den Hende, Two, and Three-Dimensional Crystallographic Fourier Summation Program for the IBM 7090 Computer, CBRL-22M-62, Massachusetts Institute of Technology, Esso Research and Engineering Go., 1962. (8) Atomic scattering factors for Bi, Br, N, and C were taken from “International Tables for X-Ray Crystallography,” Vol. 111,Kynoch Press, 1962, pp 212, 206-207, 202-203. (9) W. R. Busing, K. 0. Martin, and H. A. Levy, “ o R F L e , A FORTRAN Crystallographic Leasesquares Program,” ORNL-TM-305, Oak Ridge National Laboratory, Oak Ridge, Term., 1962. ~ vel, 111, (10) $‘International Tables for x - R ~Crystallography," Kynoch Press, 1962, PP 214-216.
(11) W. R. Busing, K. 0. Martin, and H. A. Levy, “ORFFE, A FORTRAN Crystallographic Function and Error Program,” ORNLTM-306, Oak Ridge National Laboratory, Oak Ridge, Tenn., 1964.
Volume 78, Number 9 September 1968
W. GANTMCPHERSON AND EDWARD A. MEYERS
3120
Table I11 : Atomic Coordinates of [(CH&NHg]sBiBre Obtained in the Anisotropic Refinement (Standard Deviations Are Given in Parentheses and Apply to the Last Digit of a Number; Values of Pij Listed Have Been Multiplied by 104) II
X
Bi(A)
a
Bi(B)
b a b
Br(A)
a
Br(l)(B)
b a b
Br(2)(B)
C(l)(B)
a b a b a b a b a b a
N(B)
a
C(2)(B)
a b
Br(3)(B) C(l)(A) N(A) C(2)(A)
b b
a
0 O
0 0
0 0 '/a
2
011
Bas
Baa
B12
613
623
16.6(4) 15.7(4) 14.0 (2) 13.3 (2) 20.1(6) 20.7(6) 25.2 (7) 25.8(7) 22.2(6) 22.8(6) 13.3 (5) 13.8(5) 26(5) 26(5) 25(4) 26(4) 27(5) 28(5) 26(5) 27(5) 26(4) 27(4) 34(7) 35(7)
16.6 15.7 12.9 (2) 12.2 (2) 20.3(6) 20.9(6) 21.6 (7) 22.3(7) 20.4(6) 21.0(6) 24.0(6) 24.7(7) 26 26 25 26 27 28 26 27 26 27 34 35
139 131 132 (6) 121 (6) 175(11) lQO(12) 270 (14) 284(14) lSO(12) 172(12) 194 (11) 209(12) 230 236 227 237 244 253 232 243 237 247 302 317
8.3 7.9 6 . 4 (2) 6 . 0 (2) 10.1(5) 10.4(5) 13.9 (6) 14.2(6) 10.8(5) 11.1(5) 9 . 1 (5) 9.4(5) 13 13 13
0 0 0 . 8 (8) 0 . 8 (8) -17.1(18) -17.5(18) 12.1 (22) 12.4(22) -4.5(18) -4.8(19) l.5(16) 1 . 5 (16) 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0.5 (7) 0.5 (7) -13.2(18) -13.4(19) 30.3 (21) 30.9(21) -17.8(19) -l8.0(19) -1.2(19) -1.1 (19) 0 0 0 0 0 0 0
1/6
1/3
'/e
'/a
'/3
0.0852(1) 0.0852(1) 0.1487 (1) 0.1487(1) 0.1434(1) 0.1434(1) 0.0579 (1) 0.0579(1) 0.021 (1) 0.021(1) 0.036(1) 0.036(1) O.OOO(2) O.OOO(2) 0.431 (1) 0.431 (1) 0.376(1) 0.376(1) 0.334(2) 0.334(2)
0.0763(1) 0.0764(1) 0.2523 (1) 0.2523(1) 0.2643(1) 0.2643(1) 0.3016 (1) 0.3016(1) 0.140(1) 0.140(1) 0.158(1) 0.157(1) 0.162(2) 0.162(2) 0.124(1) 0.124(1) O.llS(1) O.llS(1) 0.063(2) 0.063(2)
0.1868(5) 0.1867(5) 0.5478 (6) 0.5480(6) 0.0673(5) 0.0673(6) 0.3614(5) 0.3614(5) 0.417(6) 0.416(6) 0.253(4) 0.252(4) 0.137(5) 0.137(5) 0.102(5) 0,102(5) 0.104(4) 0.104(4) 0.176(6) 0.176(6)
Unit weights, selected data (IFol
5 325), uncorrected
for ad.
0.079, and R2 = 0.090, respectively. The net result of these computations is to show that neither the inclusion of corrections for anomalous dispersion, the removal of intense reflections, nor the introduction of anisotropic temperature factors produces significant changes in bond distances or angles in this structure, although there is the expected apparent improvement in the estimated standard deviation that accompanies a reduction in R2. The discussion will be based primarily upon the values given in rows a and b of Table I. A table of observed and calculated structure factors may be obtained from the authors upon request.
Discussion Two crystallographically independent Bi(11I)Br~ octehedra and two crystallographically independent dimethylammonium groups are present in the structure. The Bi-Br bond distances give? in Table 111, in the b rows, varx from 2.822 to 2.852 A, with an average value of 2.840 A. The Br-Bi-Br bond angles (for Br atoms cis to one another in the octahedra) vary from 87.7 to 92.5", and thus there appear to be small but significant angular distortions from the value of 90" expected for a regular octahedron. The octahedron around Bi(A) is flattened in the c direction. The mean C-N distance is 1.56 A, and the mean C-N-C bond angle is 118", with no significant deviations from these values. The van der Waals contacts are normal, greater than 3.6 A, except for Br3(B) - C2(B) = 3.47 (9) 8. There are The Journal of Physical Chemistry
13
14 14 13 14 13 14 17 18
' Unit weights, selected data (IF,l
5
0 0 0 0 0
325), corrected for ad.
several short N. .Br contacts: E(A)-Br(A) = 3.44(7) N(A)-Br(2)(B) = 3.48 (7) A, N(B)-Br(l)(B) = 3.49 (6) A, and N(B)-Br(3)(B) = 3.44 (6) 8. The (CH3)2 NH2 group that contains N(A) links the two crystallographically different Bi(II1) Bra octahedra together through N(A)-Br(A) and N(A)-Br(2)(B). The second (CH&NH2 group is associated closely only with the octahedron centered on Bi(B) and is coupled to Br(l)(B), Br(3)(B), and less closely to Br2(B), with N(B)-Br2(B) = 3.65 (6) 8. The values of the van der Waals and Ne .Br contacts fluctuate considerably in the several refinements. Schematic diagrams of the structure are given in Figures 1 and 2. The total BiBr bond order for a Bi(11I)Br~octahedron was estimated as 2.7 from Pauling's equation3 d, = dl - 0.6 log n, where dl = 2.64 8 and d, = 2.84 (average). This is lower than the values estimated for the total BiBr bond order, 2.9, obtained from the Bi(III)Br4and Bi(III)Br6structures. If d, were 2.81 A, the bond order would be 3.0, so that the expansion of the octahedron here is quite small. The individual BiBr bond orders are 0.45. The expansion may be associated with the increased charge of the ion. The corrections for the Bi-Br bond distances for the "ridingmotion" model12 were calculated, for rows a and b of Table 111, and varied between 0.01 and 0.02 A. The coupling between octahedra through N(A), and with
A,
(12) W. R. Busing and H. A. Levy, Acta Crystallogr., 17, 142 (1964).
THECRYSTAL STRUCTURES OF BISMUTH HALIDECOMPLEX SALTS
3121
Figure 1. Projection of the structure of ((CH,),NH,),BiBr, along (Wl): larger circles, Bi atoms; smaller circles, Br atoms.
../I
Figure 2. Projection of the structure of ((CH&NH&BiBra along (001): larger circles, N atoms; smaller circles, C atoms.
the N(B)-containing (CH&NH2 group,, makes the correction uncertain. The angular distortions of the octahedra are small, but, on the basis of the estimated standard deviations (0.2-0.3”). .. amear to be real. In a recent studvla of (NH&SbBn, it was found that the Sb(V)Br6 octahedron was distorted. whereas the Sbf1II)Br. mouD w&s undistorted. It may be that the group V halide anions are, in general, rather easily distorted and that
the unsymmetrically charged organic units in the present structure are responsible for the distortions observed. It is also possible that a “lone-pair” effect upon coordination may be responsible for the distortion in Bi(III)Bre. . .
__
~
I
-
Y
Acknowledgments. The financial support of The .
(13) 5. L. Lawton and
R. A. Jaoobsen,Z-V. Volume 7% Nu*
chm.. 5,743 (1966). 0
Swtanba 1068
ARVINS. QUIST AND WILLIAM L. MARSHALL
3122 Robert A. Welch Foundation is gratefully acknowledged. The facilities of the Data Processing Center of the Texas A & M University System have been used
extensively in this research. Dr. Roger D. Whealy has kindly supplied us with a sample of the compound studied.
Ionization Equilibria in Ammonia-Water Solutions to 700" and to 4000 Bars of Pressure1 by Arvin S. Quist and William L. Marshall Reactor Chemistry Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 87830 (Received December 6,1967)
The electrical conductances of 0.0100 and 0.0501 m aqueous ammonia solutions were measured to 800" and 4000 bars. Measurements are also reported for 0.0098 m NaOH solutions to 300°, together with estimates of Ao(NaOH)over the same temperature range. From the measurements on the ammonia solutions and estimates of the limiting equivalent conductances of ammonium hydroxide, conventional equilibrium constants for the hydrolysis of ammonia were calculated. From these values and their isothermal variation with the concentration of water, the complete constants, KO,were obtained that are independent of changes in dielectric constant or in density.
Introduction A convenient method for studying equilibria involving ions in aqueous electrolyte solutions at supercritical temperatures and pressures is the measurement of their electrical conductances. Equilibrium constants for ionic dissociation reactions have been calculated from measurements of this kind over wide ranges of temperature and density. Recent studies in our laboratory have included those of NaC1,2aNaBr,2band HBreS This present paper gives conductance measurements on 0.01 and 0.05 m solutions of ammonia to 700". Measurements were also performed a t 800", but the conductances a t this temperature were essentially zero even at 4000 bars. From these measurements and with estimates for the limiting equivalent conductance of NH4+ OH- at several temperatures and densities, conventional equilibrium constants for the hydrolysis of ammonia were calculated to 700". The present paper also includes some measurements on 0.0098 m NaOH solutions to 300". From these measurements, estimates were made of the limiting equivalent conductances of NaOH as a function of density to 300". By using these limiting conductances along with assumptions based on the previously observed behavior of other strong electrolytes at high temperatures and pressures, estimates were made of the limiting equivalent conductance of NaOH to 800".
+
Experimental Section The equipment and procedures used for these meaThe Journal of Physical Chemistry
surements have been described previously.2. All conductance measurements were made with the cell containing no pressure seals in the high-temperature region. A stock solution of approximately 1 m ammonia was prepared from reagent grade ammonium hydroxide (J. T. Baker Chemical Co., Phillipsburg, N. J., 30% NHs) and conductivity water. This stock solution was standardized, by using weight buret techniques, against potassium acid phthalate. From the stock solution, 0.0100 and 0.0501 m ammonia solutions were prepared and their conductances were measured to 800" and 4000 bars. A 0.0098 m NaOH solution was prepared from a standard 1.0 N NaOH solution (Fisher Scientific Co., Fair Lawn, N. J.) and standardized in the same manner as described for the stock solution of ammonium hydroxide. Reliable measurements on the NaOH solution were obtained only a t temperatures below 300". At 400" and above, the solution concentration changed rapidly because of the reaction of NaOH with the AlzOainsulation tube in the high-temperature region of the cell. Thorough flushing of the conductance cell was carried out a t the temperature and pressure of the experiment, but even then reliable values of conductances could not be obtained above 300". (1) Research sponsored by the U. S. Atomic Energy Commission under contracrt with Union Carbide Corporation. (2)(a) A. S. Quist and W. L. Marshall, J . Phya. Chem., 72, 684 (1968); (b) A. 8. Quist and W. L. Marshall, ibid., 72, 2100 (1968). (3) A. 9. Quist and W. L. Marshall, ibid., 72, 1646 (1968).
