SOMESODIUM-LITHIUMSILICA SYSTEMS
Oct., 19.79
ment of these projections by familiar methods of successive approximations lead to the complete determination of the structure (Figs. 1 and 2). The parameter data of Table I11 account satisfactorily for the intensities of X-ray reflections (Table 11) and lead to generally reasonable interatomic distances (Table IV). The structure (Fig. 2 ) is an aggregate of K+ ions, M O ( C N ) ~ -ions, ~ and water molecules.
[CONTRIBUTION FROM
THE
2863
The configuration of the M o ( C N ) ~group (Fig. 3) seems to be, stereochemically very reasonable. The Mo(CN)s polyhedron is a duodecahedron with eight vertices and triangular faces. Although required by the space group to possess only a plane of symmetry, the complex ion in the crystal approximates to the symmetry of the point group D i 4 2 m . ITHACA, N. Y.
RECEIVED JULY 20, 1939
GEOPHYSICAL LABORATORY, CARNEGIE INSTITUTION OF WASHINGTON]
Phase Equilibrium Relations in the System,Na2Si0,-Li2Si0,-Si0, * BY F. C.~
C
E
K
by both heating and cooling curve methods in studying transition phenomena.2 Homogeneous glasses of the various compositions in the system were prepared from purified silica, used in the form of cristobalite derived from powdered quartz heated a t 1500°, and from reagent grade lithium carbonate and sodium carbonate. Analyses proved the reagents all to be of high p ~ r i t y . The ~ preparations were made up quantitatively from the thoroughly dried reagents, special care being taken to avoid losses, either mechanical, or by volatilization. Losses by volatilization were negligible except in the lithium metasilicate region of the system, where they approached 0.1% of the weight of the preparation. They were almost entirely of lithium, and could be compensated for by small additions of lithium carbonate to the original mixture. The fusions were made in platinum crucibles. To Experimental Methods avoid attack on the crucibles by the corrosive The methods of study were in all respects melts of the easily fusible carbonate mixtures, it similar to those already made familiar through is necessary to exclude oxygen during the initial the publications of this Laboratory on phase stages of the heating, until the reaction of the equilibrium relations. The liquidus and other melt with the silica has absorbed most of the free significant temperatures were determined by alkali present. This was done by heating the quenching combined with microscopic examina- mixtures in covered crucibles over carefully tion, except in compositions high in lithium controlled M6ker burners (the more convenient metasilicate which crystallize too rapidly for the initial sintering of the mixtures by heating in quenching technique to be effective. In such electric furnaces always resulted in bad attack on cases the temperatures were established by heat- the crucibles), the temperature being kept a t the ing curve thermal analysis for the liquidus, and point of incipient fusion until most of the carbon * Presented before the Division of Physical and Inorganic Chem- dioxide was driven off. The final heatings were
The work described in this paper concerns itself with the general equilibrium relations a t one atmosphere pressure in the system NaSiOaLizSiO3-SiO2, and comprises the experimental determination of the temperatures a t which the various crystalline phases are in equilibrium with one another, and with melts of the appropriate composition. The crystallization relations are comparatively simple, being only slightly complicated by the occurrence of solid solutions arising from a partial mutual replacement of sodium and lithium in the silicate compounds. Only one ternary compound is present in the system. This is NaLiSi03, formed by replacement of Li for Na in NasSiO3; it is the Li-rich limit of the solid solution series (Na2,NaLi)Si03. The other silicates in the system, Li&3iOa, NazSiz06,and LizSiz06, all form solid solutions of only limited extent.
istry at the 98th meeting of the American Chemical Society, Boston, Mass., September, 1939. (1) E. S.Shepherd,0.A. Rankin, and F. E. Wright, Am. J . Sci.,
PI), 293 (lQOQ).
(2) F. C. Kracek, N. L. Bowen and 0.W. Morey, J . Phys. Chcm., 41, 1183 (1937). (3) See F. C. Kracek, Tlrra JOTJXNAL, 61, 2157 (1939), for snalyrea of reagent# and further experimental details.
Vol. 61
F. C. KRACEK
2864
Fig. 1.-Ternary diagram of the system, Na#iOpLi&iOsSiOl, showing the equilibrium fields at the liquidus, and location of the experimental points. Black dots represent compositions for which refractive indices of glasses are included in Table I.
made in electric furnaces in the usual manner, keeping the temperature as low as possible, consistent with complete displacement of the carbon dioxide. The more alkali-rich glasses were crystallized several times during the process, to drive off the carbon dioxide at lower temperatures, and thus to avoid losses by volatilization. The safe maximum temperatures, a t which losses are negligible, are roughly, 1100’ in the metasilicate, and 1300” in the disilicate regions of the system. The temperatures were measured by means of against platinum-lO~o rhodium therts, using a shielded Feussner potentiometer system. The readings of the calibrated thermoelements were converted into temperatures on the Day and Sosman gas thermometer scale with the help of standard tables.4 The temperatures of the quenching furnaces were controlled automatically with a precision of *0.3 to +0.5”. The time of the individual runs, several of which are needed for each determination by the quenching method, varied from one hour to several days, depending upon the t h which equilibrium is reached in the regions of the system. This is most and slowest for quartz The rate of dissolution of quartz increases rapidly with even small additions of lithium and, within the system, the
.
(4) L. K Adam? “Intkrnational Critical Tables,” 5‘01 I , 1926,p. 57.
reactivities of quartz and tridymite are sufficiently high to permit their mutual conversion (quartz tridymite) near the inversion temperature in about twenty-four hours. This has made it possible to redetermine the inversion temperature with somewhat less uncertainty than before. I n presenting the experimental results below, certain minor revisions of previously published data are made. Such revisions are primarily based on very frequent recalibrations of the thermoelements, combined with a general improvement both in apparatus and technique (see ref. 2 ) . Particular attention was given to the attainment of equilibrium in the charges. This is especially important in the cases of the more sluggish reactions, such as the mutual conversion of quartz and tridymite, and the various eutectic meltings, where certain phases tend to persist above the reaction temperatures for considerable periods of time. The measured limits of uncertainty of each determination are given by means of a * bracket, or by quoting the measured temperature limits, as the case may be.
Experimental Results The distribution of the compositions studied in the ternary system is represented in Fig. 1, which also shows the boundary curves between the fields. The experimental points are located mainly along certain sections across the ternary
SOME SODIUM-LITHIUM~ILICA SYSTEMS
Oct., 1939
2865
reactions. The temperatures given are the nearest determined temperatures, above, and below, the transformation point in question-in other words, the phase reaction temperature lies between the values given. Most of the measurements represent determinations of the liquidus temperatures; a t appropriate places in the table, however, temperatures for other phase reactions,
system, chosen so as to present a net of points which would allow for easy interpolation in deriving the liquidus surfaces for the various primary phases. These sections are numbered in the figure by Roman numerals to correspond with the numbering of the sections in Table I. This table contains the experimentally determined temperatures for the various phase
TABLEI RESULTS OF EQUILIBRIUM MEASUREMENTS IN THE SYSTEM, NazSiOa-Li~SiOa-Si02 No.
wtB.%
Measured limits Liquidus temp. NaaO
LinO
Si02
Below
I. Binary Section: A 31 32 33
10.0 15.0 18.0
30.61 28.94 27.92
3.31 4.98 5.98
66.08 66.08 66.10
34 35 36 37 38 39 40
20.0 25.0 30.0 40.0 50.0 i0.0 85.0
27.23 25.53 23.80 20.38 17.00 10.2 5.1
6.64 8.30 9.95 13.23 16.50 23.2 28.2
66.12 66.17 66.25 66.39 66.50 66.6 66.7
41 42 43
10.0 20.0 28.0
44 45
48 (5)
30.0 40.0 50.0 60.0 80.0 100.0
51 52 53
20.0 30.0 32.0
33.08 32.60 32.49
54 55 56 57 58 59 60 61
40.0 50.0 60.0 63.0 66.0 70.0 80.0 90
(16)
100.0
32.12 31.64 31.15 31.02 30.87 30.68 30.20 29.7 29.23
71 72 73 74
10.0 20.0 30.0 33.0
32.69 31.32 29.95 29.59
1.98 3.95 5.93 6.53
65.33 64.73 64.12 63.88
75
40.0
28.64
7.91
63.45
46 4i
OC.
= NazSizOs; B = LizSiO,
790 743 719 709 721 794 850 941 1011 1111h" 1164h"
792 746 720 710 727 795 852 942 1015
11. Ternary Section: A = NazSizOs; B 845 0.89 64.76 34.35 814 1.79 63.56 34.65 789 2.50 62.60 34.90 781 788 2.60 62.46 34 I94 814 61.17 3.58 35.25 59.95 841 4.47 35.58 860 5.36 58.75 35.89 891 7.15 56.34 36.50 906 53.95 37.11 8.94
111. Ternary Section: 2.82 64.10 63.17 4.23 63.00 4.51 5.64 7.05 8.45 8.88 9.30 9.87 11.28 12.6 14.06
62.24 61.31 60.40 60.10 59.83 59.45 58.52 57.7 56.71
Cryst. phases
Above
Refr. index quenched glass
NazSizOs NazSizOs Na2SipOs Na&lOs eutectic LizSiOa LizSiOa LizSiOa LizSi03 LirSiOs LizSiOa LizSiOa
+ LizSi03binary
= (theor.) NaaLizSiaOg NazSiz06 846 NarSizOs 816 NazSiz06 790 782 NazSizOs (Na2,NaLi)SiOa 790 (Naz,NaLi)SiOa 818 (Naz,NaLi)Si03 (Na2,NaLi)SiOs 842 862 (Naz,NaLi)Si03 893 (Naz,NaLi)SiOs 907 (Naz,NaLi)SiOs
+
A = NazSizOs; B = NaLiSiOa 796 800 NazSizOs 748 755 NazSinOa 742 743 (Naz,NaLi)SiOa 696 698 Ternary eutectic E1 758 (Naz,NaLi)SiOs 755 (Na2,NaLi)SiOa 769 772 790 (Na2,NaLi)SiOs 789 797 (Na2,NaLi)SiOs 796 805 LizSiOs 803 813 LizSi03 812 LizSiOs 841 838 862 Li2SiOa 860 873 LizSiOa 872
IV. Ternary Section: A = NazSizOs; B 823 770 708 711 696 774
= (theor.) NapLi4SiaO~ NazSiz06 825 NaZSi206 772 709 NazSizOb 712 LizSiOs 697 Ternary eutectic E1 778 LilSiOs
1,513 1,517 1,521
1.521 ...
1,527 ...
... ... ...
1,509
... ...
... 1.521 1,523 1.527 1.528 1.514 1,517 1,519
... 1,525 1.528 1.531 >1.530
... >1.530 1.535 1.538 1.511 1.516 1.521 1.523 1,525
TABLE I (Continued) Measured limits N o " I
i 1)
r-
i ,
78
79 (211
Liquidus temp.
Wt. % B
NazO
1,izO
Si02
Below
50 0 60.0 70 0 x c5 100
27.28 25.94 24. 59 22.5 20.5
9.89 11.87 13.85 16.8 19.8
62.83 62.19 61.56 60.7 59.7
844 889 927 977
5 0 10.0 20.0 24.0 29.2
32 29 30.64 27 23 25.87 24 11
0.99 1.99 3.98 4.78 .i.Xl
ti6.71 67.37 68.79 69.35 70.08
86
33.0
23.15
6.37
70.48
87 88 89 90
40.0 50 60.0 70.0 80.0 do. 0 100.0
20.43 17.19 13.62 10.20 6.81 3.41
7.97 9.94 11.95 13.93 15.92 17.90 19.92
71.61 72.87 74.43 75.87 77.28 78.69 80.08
101 102 103 104 105 106 107 108 109 110
10.0 20. (J 25 0 32.0 :34,0 36 0 38 0 10 0 42 CI 50 0
11 1 112
80 0
45. 70 40.62 38 10 34 54 33 52 32 19 31.48 30.45 29 45 25.36 20 26 10 14
81
91 112
ti0
123
..
121
.. ..
131
..
9 97
13'7 133
. .
134 135
.. ..
10 00 10 00 9.98 9 98
141 142 143
..
15. 00
..
144
..
..
847 893 933 981
Cryst. phases
LipSiO3 LiZSiO, LizSiO, LizSi08 Li2SiOa
14.98 14.98 14.99
7.99
77.02
.
= NapSizOa;B = LipSizOa
845 816 758 735 701 692 711 692 780 849 908 954 986 1012
846 818 762 736 702 694 712 695 782 851 912 956 989 1014 1034c
Na2Siz05 NazSi2Oa NapSip05 NazSi20a Na2Siz06 NazSi2Os Li&i03 LizSi03 NaZSizOs 4- Li~Si03 LizSiOl LirSiO:, LipSiOa LizSi03 Li&Oa LizSiOa Li2Si08
+
+ LinSizOa
VIII. Ternary Section a t 10.0% NazO 1079 Tridymite 81.44 1075 8.59 958 Tridymite 954 10.00 80.00 906 RL18 LizSi20~ 11.01 78.99 914 9 1fj Li*Si2Oj 11.97 78.05 11154 1056, Li2SiOa 18.96 71.06
Ternary 80.49 78.52 77.33
glass
,
VI. Ternary Section: A = NazSi03; B = Li~Siz06 1037 (Naz,NaLi)Si03 1032 1.99 32.31 (Na2,NaLi)SiOa 956 958 3.98 55.40 (Na2,NaLi)SiOa 906 908 4.98 50.92 (Na2,NaLi)SiOl 838 840 6.37 59.09 (Na2,NaLi)SiOa 819 821 5.77 59.70 (Kaz,NaLi)Si03 800 803 7.14 li0.37 (Nap,NaLi)Si03 777 783 7.57 60.95 762 761 LizSi03 7.92 61.62 780 783 LipSi03 8.34 62.21 818 850 LipSiOa 9.91 04.73 911 913 LiaSiOB 11.88 67.86 996 998 LilSi03 15.90 73.96
IX. 4.51 6.49 7.69
Relr index quenched
I.529
W I . Ternary Section a t 3.070 Nan0 11.30 82.71 1088 1092 Tridymite 19.46 81.55 999 1001 Tridymite 1012 1019 Tridymite 15.96 81.04 1002 1004 IdzSi20a 16.46 8i).55 1004 1005 LizSip05 17.46 79.55 1003 1106 Li2SiOa 21.94 73.OG 1089 1091 LizSiOn 25.95 71 0*5 1146h Li&03
125 126 127
. .
Above
l00lL
Y. 'Ternary Section: A 82 83 84 85
GC.
Section at 15.0% Na20 1075 1080 Tridymite 910 Tridymite 904 811 Quartz 805 796 799 Quartz LilSizOs 802 804 LilSiOa, (LisSilOs)
+
1.507 1.509 ...
1 . 513 1.515 1 515 1.518
... 1.526 1.530 1.532
... 1.519 1 .523 1.523 ...
>1.525 ...
1.528 1,532 1.519 1 ,524 1 528
1,509 1.518 1.521 1.530 1.501 1.510 1,513 1.515
SOME SODIUM-LITHIUM~ILICA SYSTEMS
Oct., 1939
2867
TABLE I (Continued) NO.
Wt.% B
NalO
Liz0
Si02
X.
Measured limits Liquid,”” temp. C. Below Above
Refr. index quenched glass
Cryst. phases
Ternary Section a t 20.0% NazO 78.03 974 980 Tridymite 77.0 871 874 Tridymite 76.52 829 833 Quartz 75.02 688 692 Quartz 642 645 Quartz f NazSi20s 636 638 Ternary eutectic E2 74.42 660 663 LipSiOs 637 639 Ternary eutectic Ea 73.51 708 709 LizSiOs 70.00 854 856 LinSiOa
151 152 153 154
..
.. .. ..
19.97 20.0 19.98 19.98
2.00 3.0 3.50 5.00
155
..
20.28
5.30
156 157
.. ..
20.00 20.00
6.49 10.00,
161 162 163
.. .. ..
21.99 21.96 22.00
2.00 3.00 3.70
76.01 75.04 74.30
164 165 166 167
.. .. .. ..
21.95 21.96 22.00 21.99
3.99 5.00 5.59 5.98
74.05 73.04 72.41 72.03
171 172 173 174 175
.. .. ..
25.00 24.99 24.99 25.00 25.00
1.00 2.00 4.00 6.00 6.50
32.50 32.47 32.47 32.46 32.40 32.48 32.45
6.50 7.98 8.99 9.96 10.93 11.92 14.02
1.498 1.502 1.506
... 1,509
... ...
XI. Ternary Section a t 22.0% Na10
,
..
..
827 739 686 686 684 677 675 688
831 747 688 689 686 680 676 689
Quartz Quartz Quartz 4- Na2SirOs NazSiz06 NazSi~06 Na~SipO6 NazSi~Os LizSiOs
XII. Ternary Section at 25% NazO 74.00 766 767 Na2Sip06 73.01 762 764 Na2Si20s 71.01 735 737 NazSizOa 69.00 707 711 NazSizO6 68.50 714 716 LizSiOa
...
...
1.507
... ... ... >1.510
... 1.501
... ... 1.517 1.519
XIII. Ternary Section at 32.5% NazO 181 182 183 184 185 (9) 186
..
.. .. .. ..
.. ..
61.00 59.55 58.54 57.58 56.67 55.60 53.53
780 812 828 841 847 851 834
782 816 831 843 849 853 842 83611
(1%) 191 192 (60) 193
36.00 37.50 39.00 40.53 42.00
1.524 1.528
(Na2,NaLi)SiOs (Naz,NaLi)SiOa (Na2,NaLi)SiOS (Na2,NaLi)SiOa (Na2,NaLi)SiOa (Na2,NaLi)SiOJ (Na2,NaLi)SiOs (Naz,NaLi)SiOs
... >1.530
... ... ... ...
XIV. Miscellaneous Ternary Section from NazSi03 thiough Composition 60d 9.96 57.58 32.46 841 843 (Na2,NaLi)SiOa 31.75 10.43 57.82 832 834 (Na2,NaLi)SiOs 30.98 10.85 58.17 830 832 (Naz,NaLi)SiOs 30.20 11.28 58.52 838 841 LizSiOa 29.46 11.68 58.86 854 855 LizSiOa
XV. Miscellaneous Compositions 63.11 701 703 LizSiOs (Na2,NaLi)SiOs 719 720 (Naz,NaLi)SiOs 202 .. 30.20 6.30 63.50 696 697 Ternary eutectic E , 699 700 (Na2,NaLi)SiOs LizSiOa 707 708 (Na,,NaLi)SiOa 203 .. 17.03 10.32 72.66 863 864 LizSiO~ .. 15.29 204 7.14 77.57 828 837 Quartz 205 .. 11.49 9.68 870 Tridymite 78.83 869 868 869 Li2SiZO6 tridymite 206 .. 12.35 6.86 80.78 1052 1058 Tridymite a “h” denotes determination by heating curve analysis. Interpolated from data on NapSiOsLiBiOa, on LiBiOrSiO~. B 27,$2% LilO, 72.18% Sios. > more than. 201
..
30.60
>1.530 1.532
... >1.530 1.534
+
6.29
1.523
+
+
-
-
‘
1,523
... 1.514 1.506 From data
2868
Vol. 61 TABLEI (Colecludtd) Polymorphous Transitions in Na2SizOb Composition, wt. % LirO
NatO
34.05 37.5 35.8 31.7 31.0
25 0 30 61 28 94 :lo 64
3 31 4 9s
I 99
Transitioz temp.4 av.. C.
Si09
1-11
11-111
65 95 62,5 64 2 68.3 69.0 73.0 66.08 A6 08 ii7 37
706 8 708.7 705 9 707 1 70'7 7 707 6 699 5 698 1
677.0 677.1 673.1 678.4 677.8 Bi7.2 687.7
688.5
16%
(j88.1
ii
Remarks
Pure NazSizOs Excess NazO Excess NalO Excess Si02 Excess Si02 Excess Si02 90% NazSiz06 85% Na2SizOo 90% NatSizOs
+ 10% Li2Si03 + 15% LizSiOa
+ 10% Li2Sis0,
Polymorphous Transitions in Li&zO6
... ... ...
19.92 80 08 (96O)h' 936.6 22.1 7'7 9 935.4 17.7 82 3 (960)h' 937.2 12.6 87.4 935.5 3.0 17 5 79.5 929.5 3.4 17.9 79.7 929.8 a For method of deducing the average transition temperature see Fig. 3.
Pure LilSiOS Excess Liz0 Excess Si09 Excess Si02 90% LiSiiOs + 10% NazSiaOs ' Faint arrest on lirst heating only.
mainly eutectics, are given, in connection with 1. The System, NaaSi03-Li2Si03.-The data the composition for which they were determined. on this system have been published recently." The coordinates of the invariant points for the pres- This system is of special interest in that sodium sure of one atmosphere are collected in Table 111. metasilicate crystals are capable of exchanging sodium for lithium in solid solution until the The Binary Systems on the Side Lines composition sodium lithium (1,l) metasilicate is Before proceeding to a discussion of the rela- reached. Lithium metasilicate, on the other tions brought out in the ternary system, it is hand, takes up only a minor proportion of sodium ____-into its lattice. The compound 1200 I sodium lithium (1,l) metasilicate melts incongruently at 847 O , yield' ing a liquid of 39.3 wt. per cent. Liquid lithium metasilicate, and the so; - dium-bearing lithium metasilicate 1100 ; crystals. The solid solution series, (Naz,NaLi)SiOa, has a minimum on its liquidus at &Iso, 38.5 wt. per cent. ,ci lithium metasilicate (binary compolo00 sitions) The equilibrium relations are illustrated in Fig. 2 (see also , i I l s ~ ~Fig. 2 of ref. 3). OC?
.
'
I
ROO
ss
-'
o
L.
m CL-Lw- c(:n an 1 R~t DybLGIIl, llU2L231V3-31V2.
system has been studied by Morey and BowenJ6and by Kraceke who extended the investigation to include the orthosilicate, NarSiOl. 10 30 50 70 90 (a) Polymorphism of Sodium binary system, NazSiO~-Li&iOs. Reprinted from Ref. 3. ~ i ~ ~ supple-~ I
I
.
-This
1
Fig. 2.-The
necessary to describe the fundamental binary e, Na2SiOasystems on the sides of the t Li&Ot,--SiOz. It will be noted that one of these, Na&iO&i&iC&, is a binary section acrom the larger triangle, Na@Li@SiO*.
menting those earlier made by Kracek,6of thermal arrests in the sodium disilicate region with the improved apparatus now in use (see ref. 2 ) were made
(liiijGW. Morey snd N. L. (6)
Bowen, J . Phys. Cham., 1% 1367
F. C . Eroeek, W d , , 84, 1688 (1980).
~
Oct., 1939
SOMESODIUM-LITHIUM-SILICA SYSTEMS
2869
on pure sodium disilicate (65.95% silicon dioxide), on two preparations toward sodium metasilicate, a t 62.5 and 64.2% silicon dioxide, and on three preparations toward silicon dioxide, a t 68.3, 69.0,
660
Fig. 3.-Typical
heated completely crystalline preparations of 74.8 and 76.0% silicon dioxide a t various temperatures, and quenched. When heated for two days at 786 O, the preparations remained powdery. At 788; they sintered to porous lumps, with little or no glass evident under the microscope. At 790" there was much glass, in which were imbedded sharply faceted quartz crystals, and larger eroded quartz grains, with very little unevenly scattered remnants of sodium disilicate crystals. The eutectic temperature is accordingly 789 * 1". The quartz liquidus on the 74.8% silicon dioxide preparation (compare Kracek's earlier preparation 571, 74.81% silicon dioxide, 838 ") was found by similar procedure t o be 822 * 4". The intersection of the liquidus curves for sodium I disilicate and quartz, a t the newly determined eutectic temperature of 789", is 680 700 720 Temperature, "C. located a t 74.2 =t0.3% silicon dioxide. differential heating and c d h g curves of phase The former values for the eutectic were 793", and 73.4% (Morey and Bowen) and 73.9% (Kracek) silicon dioxide. (c) The Na2SiO;-Na2Siz05 Eutectic.-Precrystallized preparations of 61.5 and 62.5% silicon
transitions NazSirOs, giving geometrical construction for deducing the average in transition temperatures.
and 73.0% silicon dioxide. Each of these preparations yielded two arrests, one a t 678", the other at 707" (average temperatures), as illustrated in Figs. 3 and 4. The heat effects are relatively small, probably of the order of 2 cal./g. for the 678" 100 arrest, and 1 cal./g. for the 707" arrest. The 707" arrest had escaped detection previously in preparations containing excess silicon dioxide. An arrest, not included in Fig. 4, previously re901 corded a t 768 " in silicon dioxide-rich preparations only, is again recorded in the same region, both of temperature and composition, and probably is ti due to eutectic melting of small amounts of triu dymite and cristobalite present. These measurements lead to the conclusion 801 that sodium disilicate is trimorphous with inversions located at atmospheric pressure at the average temperatures of 678 and 707'. Presence of lithium silicates influences these temperatures, as stated a t a later point in the paper (see Figs. 8 701 and 9). (b) The Na2Si20s+mrtz Eutectic.-The determination of the NazSiz05-quartz eutectic by heating curves is inexact, yielding an arrest which is spread rather over-the temperature Fig. 4,-The disilicate region of the system, Nasiosinterval from about 788 to 8oo"* To SOo: open circles by Morey and B ~ w e n ,and ~ Bowen, verify the location of this eutectic, we have Schairer and Willerne (unpublished); black dots, Kracek. 0
2870
F. C. KRACEK
dioxide remained unsintered at 8.36") and contained much glass at S38". If 837 * 1" is taken as the eutectic temperature, the eutectic composition is 62.1 * 0.2% silicon dioxide. The former values were S40", 60.8% silicon dioxide (Morey and Bowen), and 8-1-6O, 02.17; silicon dioxide (Kracek). (d) The Cristobalite Liquidus.-The liquidus temperatures published earlier by the writer €or cristobalite in the system, NazOSiOz,are slightly higher than those determined in the present work. The high values in the earlier determinations can be ascribed only to the existence of a temperature gradient in the quenching furnace then used. The newly determined values are: 1532 * 5" at 91.5, 1580 * 3" at 95.0, and 1628 * 3" at 97.5% silicon dioxide. 1100
1050
5
.. 1000
950
I 75
80
SiOz, wt. %.
8.5
Vol. 61
Jaeger and van Klooster8and by Kracek, 7,9 among others. (a) The Cristobdite Liquidus.-A redetermination was made of the liquidus temperatures for cristobalite as the primary phase. It was lound that the published temperatures7 in this system also are slightly high; the new values are 1508 * 1"a t 92.5, 1361 * 3" a t 95.0, and 1605 * :i" a t 97.0% silicon dioxide. (b) Polymorphism of Lithium Disilicate.To verify the formerYthermal analysis measuremerits in the lithium disilicate region of the system, we have made new determinations on crystallized preparations of 77.9, 80.08 (LizSi206), 82.3, and 87.4% silicon dioxide. A small arrest, with a thermal effect of 1 to 2 cal./g. in pure lithium disilicate, was found in all the preparations, at the average temperature of 936'. This arrest has the characteristics associated with typical polymorphic inversions, with hysteresis. The very small irregularity in differential temperature, formerly noted at 960", could not be reproduced with certainty. It was noted once, but only during the first heating, in the 80.08 and S2.3y0 silicon dioxide preparations. It did not reappear during subsequent heat treatments, or in any of the other preparations; hence, its significance remains unexplained. Addition of sodium silicates to lithium disilicate lowers the 936" inversion to 930" (see the section, Na8i206-Li~Si205). The phase relations in the neighborhood of lithium disilicate for the system LizSiOrSiOz are illustrated in Fig. 5. The Ternary System The only ternary compound in the system is sodium lithium (1,l) metasilicate. This compound is an end-member of the (Na2,NaLi)SiOa series of solid solutions, and has no separate field of its own in the ternary system. Hence, the
Oct., 1939
SOME
SODIUM-LITHIUM-SILICA SYSTEMS
287 1
/
Fig. 6.-Equilibrium diagram with isotherms, of the system, Na2SiOa-Li9iOs-SiOa. Open circles represent interpolated temperatures corresponding t o the various isotherms. Black dots give the location of boundary curves as measured or deduced by interpolation from the data on the various cross-sections in the system.
is partly ternary a t the liquidus; this is due to the incongruent melting relations of lithium disilicate. Method of Interpolation in Deriving Isotherms.-The data for each of the sections I to VI across the system, as indicated in Fig. 1, were plotted on a sufficiently large scale to allow for interpolation with a precision of k0.5". The compositions corresponding to even temperatures a t 50" intervals were then read off from the graphs, and plotted in the ternary system, as indicated in Fig. 6 by the open circles. Cross interpolation was employed for the remaining sections, VI1 to XIV. For this purpose the temperatures of intersection of these sections with the sections I to VI were read off from the above graphs, and were then plotted, together with the data for preparations on the individual sections, as indicated in Fig. 7, the abscissa being progressively shifted so as to bring the metasilicate composition for each of the sections VI1 to XI11 upon the vertical line a t the left of the diagram. This was done by adding to the lithium oxide content for each section the quantity % Liz0 added
=
% NarO along the section X
Liz0 in LiaSiOa NaaO in Na2SiOa = 0.6541 X % NarO along the section
The resulting figure, Fig. 7, represents a t-x (temperature-composition) projection of a part of the system upon the plane of the side system, LizSiO&Oz. To avoid confusion in the diagram, only sections up to 25% sodium oxide are represented. The sections I, V, and VI, appear as curved dotted lines, with the interpolated loci of the intersections marked by crosses. Other t-x projections of the system were constructed to facilitate cross-interpolations in regions not covered by Fig. 7. The data so obtained are plotted in Fig. 6. The Boundary Curves.-The coordinates of the boundary curves between the various primary phases, as deduced by interpolation from the graphs, or directly from the experimental data, are given in Table 11. The boundary curves meet a t the invariant points, which are collected for reference in Table 111. 1. The System, Na2SizO~-Li2SiO3.-The relations in this system are illustrated in Fig. 8. The system is simply eutectoid ; both the end-members are present in limited solid solution. The extent of the solid solution in lithium metasilicate is approximately the same as in the system NazSiOr LizSiOs, as judged from the values of the refractive indices of the crystals. The solid solution
P. C. KRACEK
Vnl. 61
TABLE I1 C o o ~ n r s a r ~OFs THE BOYWARYCURVESIN THE SYSTEMNa2SiO3-Li2SiO3-Si0z Sectioo no. arid identification
I1 I11 XI11 ... "..
.,. XI\? I11 VI ... ...
... IV I XI1 V
XI
...
NarO
Wt. ?Gf Liz0
d. Boundary between (Na&a,Li)SiOa and I\;arSizOh (s. s.1'' 37.9 .. Na~SiO3-XazSizO; 35.0 2.7 NazSizOfi-(NarLiZSirOsj" 32.5 4.5 Na&Os-NaLiSiO:, 32.5 4.6 32.50/, XazO 30.0 6.30 30.0% Ka:O 29.9 6.2 Ternary eutectic E: B. Boundary between (Nan,hTaLi)SiOsand LizSiOa (s. s.) 30.8 13.1 Ka?SiOs-Li&5iOs xa2Si03-72.19% si02 on LinSiOa-SiO? 30.7 11.0 30.8 9 .5 Na&?&O~-i\'aLiSiOB 30.6 7.9 KakXOaLipSi20j 30.0 6.4 30.0% XaeO 29.9 6.2 Ternary eutectic E1 C. Boundary between NazSi20s (s. s.) and LipSiOa (s. s.j 29.9 6.2 Ternary eutectic E1 29.7 6.3 NazSipOa(NazLi4Si30s) 27.5 6.4 NaSi205-Li&3O3, eutectic 25.0 0.3 25.0% NasO 23.7 6.0 Na&2ObLizSizOj 22.0 5.7 22.0% Nan0 20.5 5.2 Quint. point Qi
Si02
1. o c .
62.1 62.3 63.0 62.9 63.7 63.9
837 * 1 782 741 74 1 700 697 * 3
56.1 58.3 59.7 61.5 63.6 63.9
847 * 1 831 799 759 700 697 * 3
63.9 64.0 66.1 68.7 70.3 72.3 74.3
697 698 709 705 693 674 641
*3
5.2 5.1
74.3 74.4
641 637
*3 *
3
1:. Boundary between NazSiz06 (s. s.) and Quartz 25.8 ... NanSiz06-Si0. 25.0 0.8 25.0% NarO 22.0 3.6 22.0% Nap0 20.5 5.1 Ternary eutectic E,
74.2 74.2 74.4 74.4
789 770 688 637
*
1
74.3 74.6 77.0 78.0 79.4 80.1
641 660 803 915 1008 1033
*
3
*
1
82.2 81.3 79.2 78.6
1028 * 1 1001 904 867 * 3
78.6 77.2 74.7 74.4
867 = 3 801 657 637 * 3
75.5 76.2 76.9 78.0
867 * 3 867 867 867
78 6
867
*3
*3
I). Boundary between NanSipOa (s. s.) and LizSizO6 (s. s.)
... ...
...
XI1 XI ...
... X IX VI11 VI1 .
.
I
... VI1 VI11
... ... IX X ..
... XI
X IX
Quint. point Q1 Ternary eutectic E ,
20.5 20.5
G. Boundary between Li2SiOB(s. s.) and Li&i208 (s. s.) 20.5 5.2 Quint. point QI 20.0 5.4 20.0% N a 2 0 15.0 8.U 15.0% Na20 10.0 12.0 10.Oyo"azo 3.0 17.6 :3.0% NasO .. 19.9 Li,SiO~-SiO? €I. Boundary between Li&iz05 (s. s.) and Tridymite Li2SinOh-SiO:: 17.8 3.0% NaaO 3.0 15.7 10.0% Nap0 10.0 10.8 11.9 9.5 Quint. point Q: K, Boundary between LizSi206 (s. s.) and Quartz 11.9 9.5 Quint. point Q z 15.0% NaeO 15.0 7.8 20.0 5.3 20.0% Na2O Ternary eutectic E? 20.5 5.1 I,. Boundary for Inversion of Quartz to Tridymite ... NazSiOp--Si02 24.5 22.0 1.8 22.0% NazO 20.0 3.1 20.0% Nap0 15.0 7.0 15.0% NazO 11.9 $1 . 9 Quint. point.
*3
Oct., 1939
SOME SODIUM-LITHIUMSILICA SYSTEMS
2873
TABLE I11 INVARIANT POINTSFOR THE PRESWREOF ONEATMOSPHERE IN THE SYSTEM Na&iOTLiSiO&Oa
NaLiSiOa and LizSi03 (Na2,NaLi)SiOa NazSiOs and NazSi~Os NazSi~O5and quartz LinSinOaand tridymite N a S i ~ 0 5and LizSiOB (Naz,NaLi)Si03,LizSi03,and Na2Siz05 NazSizO5, LizSinOa,and quartz NazSi20a,LizSilO5, and LizSiOa LinSinOj,quartz, and tridymite Quartz and tridymite XazSiz05 I and NazSiZOj I1 NazSiSOb I and Na2SizOs I1 in solid solutions NazSi20r I1 and NazSipOs I11 NaLSizOs I1 and Na2Si206 I11 in solid solutions LizSiz05I1 and Li~Si205111 Li2Siz05I1 and LizSi205I11 in solid solutions
Type
NarO
Lis0
5i01
Melting Melting Melting Peritectic Liquidus Peritectic Minimum Eutectic (binary) Eutectic (binary) Eutectic (binary) Eutectic (binary) Eutectic (ternary) E1 Eutectic (ternary) EZ Peritectic limit, Q1 Inversion point, Qz Inversion Inversion Inversion Inversion Inversion Inversion Inversion
50.79 34.04
... ...
49.21 65.96 66.78 80.1 80.08 56.13 55.96 62.1 74.2 82.2 66.1 63.9 74.4 74.3 78.6 100.0 69.56
...
... ... 30.82 31.25 37.9 25.8
... 27.5 29.9 20.5 20.5 11.9
33.22 19.9 19.92 13.05 12.79
... ... 17.8 6.4 6.2 5.1 5.2 9.5
34.04
... ... ... ...
...
...
...
...
19.92
80.08
...
...
I
.
.
34.04
...
...
65.96
1.
oc.
1089 874 1201 1033 1034 847 845 837 789 1028 709 697 637 641 867 867 707 699 678 688 936 930
*1 *3 *1 *1 *1 * * * * *
1 1 1 1 1
*3 *3
*3 *3
*3 *3 * *
* *
*
*
3 3 3 3 3 3
1700 in sodium disilicate is approximately 3 * 3y0 lithium metasilicate, as indicated in the diagram; the 678" transition in sodium disilicate is raised to 688", while the i07" transition is lowered to 1500 699") as measured by thermal analysis on preparations 31 and 32 (see Table I). The binary eutectic is a t 709 * 3", 19.20% lithium metasilicate in the bi1300 n a b system (6.38% lithium oxide, 27.50% sodium oxide, 66.12% silicon dioxide in ternary compositions). 2. The Section, Na2Si205-Li2Si205. P u -The phase relations along this section 1100 are only partly binary a t the liquidus. Lithium disilicate melts incongruently a t 1033 * l o , the primary phase a t the liquidus, 1034 * 1", being lithium metasilicate. Additions of sodium disilicate 900 to lithium disilicate lower the peritectic temperature more rapidly than the liquidus, as shown in Fig. 9, so that lithium metasilicate continues to be the 700 primary phase down to the boundary with sodium disilicate, 693 '. Beyond 30 20 . 10 this boundary the liquidus relations are Fig. 7.-Temperature-composition projection of part of the system binary, sodium disilicate being the priupon the t-x plane of the system, LizSiOsSiO2. Open dots are exmary phase, but the sub-liquidus rela- perimental points. Crosses give intersections of the sections VI1 to tions remain ternary to 641 * 3", a t XI11 with sections I. V and VI. Dot and dash curves reuresent which temperature the last of the lith- phase boundaries, and I h r k circles, interwctions of phaw houndariei.
F.C.KRACEK
28i-1
40 60 80 LinSiOj,wt. %. Fig. 8.-The binary system, Na2Si?Oj-Li2SiOs. Solid solution relations, as estimated, are indicated by dotted lines. a3
20
VOl. 61
sodium tungstate used by Fenner. Van Nieuwenburg" reports the conversicn of quartz to tridyniitc in the presence of 1%;. of lithium carbonate, but he apparently has made no effort to measure the equilibrium temperature closer than about *30°. Since carbon dioxide is displaced rapidly from such mixtures, the action depends 011 the speed of reaction between the quartz and the small quantity of crystalline lithium disilicate formed. Fluxing action beconies important only above 1028' (see Fig. 4). For this reason, additions of lithium carbonate to quartz are actually less efficient than those of sodium tungstate in promoting the inversion to tridymite (unpublished experiments by the writer). However, since lithium ions, presuxnably because of their small dimensions, tend to act as efficient devitrifying agents, it is reasonable to suppose that the inversion would be accelerated by lithium-bearing melts, fluid below the inversion temperature. Compositions in the present system, in the silica field, with liquidus temperatures above 900" fulfil this requirement, and it was
ium metasilicate crystals, and liquid, disappear. The solid solution relations for sodium disilicate and lithium disilicate are indicated in Fig. 9 by dotted lines, the es'Oo0 tent in either constituent being estimated to be 5 * 3c/;. of the other, as based on measurements of the transition temperatures. As already mentioned i n discussing the hinary systetn, NazSi&:, !100 I,izSiOaJadditions of lithiurn silicates to sodium disilicate lower the 707' inver .;ion to W Y " , and raise the 678' inversion C)n the other hand, the 934" 117 688'. .iiiversioti in lithium disilicate is lowercd ROO hy additions of sodium disilicate to 930 ^. The Quartz-Tridyfnite Inversion. 'I he temperature of this sluggish, but reversible, traiisformation is given hy Fperiner as 870 * LO', from measure ments made in the presence of sodium tungstate, NazW04, as flux.'@ Many other substances, such as phosphorus pentoxide, lithium carbonate, potassium carbonate, sodium carbonate, and tung20 40 60 80 Li2SizOs,wt. %. sten trioxide, have been suggested as -fluxing agents for accelerating the inver- Fig. Q.--The ternary section, Na*Si&&-Li~Si~O5 Binary at the these, only lith- liquidus of Na&Ob, and below 641". Estimated solid solution resions in silica. bong lations are indicated by dotted lines, ium carbonate holds, a prim$, any promthan the ise Of being 'Inore efficient
'
(11) C J. van Nieuwenburg, Rcc. trao. chim. de Pays-Bas, 48,
(10) C
p\'
Fenner A m
J Sci 36,311
(l1Il'j
402 (1929), summary.
Oct., 1939
SOME SODIUM-LITHIUM~ILICA SYSTEMS
found that the most suitable were those whose liquidus lies in the range between 900 and 1000°. The melts of compositions still richer in silicon dioxide tend to be too viscous to act efficiently as fluxes. In the region mentioned (see Fig. 6 ) quartz is converted into tridymite almost completely (only the larger grains remaining undissolved, but rounded) in twenty-four hours a t 869', and in forty-eight hours a t 868'. In the reverse direction, tridymite is converted into sharply faceted crystallites of quartz in twentyfour hours a t 866", and in less than twenty-four hours at 865 O . Similar conversions were obtained with all the following compositions: nos. 131, 132, 141, 142, 151, 152. It is concluded that the inversion temperature a t one atmosphere pressure is 867 * 3', as measured by thermoelements calibrated a t the sodium chloride (800.4') and gold (1062.6') points. This confirms Fenner's temperature of 870 * 10". The temperature 867' is used in presenting the experimental results of this paper. The liquidus fields of the various primary crystalline phases are smooth sheets pointing down to two ternary eutectics E1 and Ez (Table 111). Eutectic El represents the junction of the boundaries between the fields of LizSiOa, NazSizOs and (Na~,NaLi)SiOs, a t 697 * 3'. Of these boundaries, the one between Li2Si03and (Naz,NaLi)SiOa has some theoretical interest. Within the triangle, Na2Si03-NazSi~06-NaLiSi03,this boundary is a peritectic (reqction) curve and represents the lowering of the incongruent melting point of sodium lithium (1,l) metasilicate (847" in the pure compound) with the changing compositions of the melts in equilibrium with the crystals. There is no eutectic, but only a minimum on the liquidus of the solid solutions (Na2,NaLi)Si03 in the binary system, NazSiOs-LizSiOa (see also Fig. 2 of ref. 3). As the boundary curve crosses the join, Na2SizO5-NaLiSi0sl this minimum coincides with the curve, which then takes on the character of a eutectoid trough, within the larger triangle, NazSi03-NazSizOs-Li2SiOa, and finally ends at the ternary eutectic. The boundary between sodium disilicate and lithium metasilicate, beginning a t the ternary eutectic El, 697", rises to the temperature of the binary eutectic between these two compounds, 709 * 3O, and then descends to the quintuple point Q1, 641 * 3", which is the lowest tempera-
2875
ture on the peritectic curve for the incongruent melting relations between lithium disilicate and lithium metasilicate. The ternary eutectic Ez,a t 637 * 3', practically coincides with the quintuple point QI; similarly, the two boundary curves, one between lithium disilicate and lithium metasilicate, the other between lithium disilicate and quartz, lie very close to each other for a considerable portion of their lengths. However, they cannot coincide, since the ternary eutectic EZ is located within the triangle, NazSizOs-LizSi2O6-Si0 2 , so that the eutectic relations demand that lithium disilicate, and not lithium metasilicate, be one of the eutectic constituents. This will be made clear also by a study of the crystallization relations in Figs. 6 and 9.11a Refractive Indices of the Glasses Each preparation studied in the course of this work was obtained in the form of a homogeneous glass. A measurement was made of the refractive index of the glass of each composition represented by a black dot in Fig. 1. These measurements were made on crushed, quenched glasses only, by the immersion method, and are estimated to be accurate to *0.003. The measured values are recorded in Table I, and values along several of the ternary sections in the system are plotted in Fig. 10. This figure also includes published values for LizO-SiOn glasses19and an interpolated curve for annealed NazO-SiOz glasses.12,13 Cocrystallization of Sodium and Lithium Compounds in Solid Solutions.-Mutual replacement of sodium and lithium plays a sufficiently important part in the crystalline phases of this system (see Figs. 2, 8, 9) to make it desirable to inquire into the general problem of solid solution formation between other sodium and lithium compounds, particularly since it appears to be held by some writers that no such solid solutions have been establishedI4 for these and other pairs of compounds with cations of the helium and neon-like structures. This view is based on a consideration of the ionic radii, and the differences between the structures of the cations. (Ila) The principles involved in this and the immediately preceding paragraph are discussed in standard books on heterogeneous equilibria, e. g . , Rudolf Vogel, "Die heterogenen Gleichgcwichte," Akademische Verlagsgesellschaft m. b. H., Leipizig, 1937, pp. 410 el seq. (12) C. A. Faick and A. N. Finn, Bur. Slandards J . Research, 6, 993 (1931). (13) G. W. Morey and H . E. Merwin, J . Optical Soc. A m . , 31, 832 (1932). (14) See H. G . Grimm and H. Wolff in "Handb. der Physik," 94, part 2, Verlag von Julius Springer, Berlin, 1933, p. 1097, Table 78.
2876
F.
c. KRACBR
VOI. fil
also would present a complete series of solid solutions. In the chlorides unmixing occurs below the critical temperature, 2'70-300 O, 38 mole 1.550 per cent. lithium chloride, the minimum on the liquidus being 553", 72 mole per cent. lithium chloride; the fluorides may thus be expected to yield continuous solid solutions with unmixing d very near the minimum on the liquidus, or actually 1.500 above it, that is, two series with eutectoid relations. The carboxiates,molybdates and tungstates are especially interesting. The sodium compounds are all three, at least partially, miscible with sodium sulfate; among the corresponding lithium compounds only the carbonate is known 50 60 io 80 90 to be miscible with lithium sulfate; the tungstate Si9, wt. %. and molybdate have not yet been investigated, Fig. 10.-Refractive indices of quenched glasses in and crystal structure information is also lacking. . identithe system, N ~ S i ~ - L i & i W i 0 2Numbers It is highly probable that miscibilities analogous fying the curves denote percentage content of NanO. interpolated from data of The curve for N&iO&iOz to those in the sodium compounds will be found, Faidr and Finn,and of Morey and Merwin, on anand that, consequently, small amounts of solid nealed glasses. solution will be found also, by sufficiently sensiSodium and lithium compounds probably present tive methods, in the corresponding systems of n to the rule, and, for that rea- lithium and sodium compounds. The data on f the known experimental data these systems, as they stand, are decidedly insufficient to deny the existence of such small is desirable a t this point. The existence of solid solutions in systems of amounts of solid solutions as exist between crysds having the same talline sodium and lithium disilicates (Fig. 9); sodium and lithium e following systems nn the other hand, it is certain that the miscibilanion is have been studied: LiC1-NaC1,l5 LiEr-NaBr,l6 ity, if it exists, must be very limited. One further case requires consideration. The L&AlFe-Na&Fs, LiNOa-NaNO;t, l8 LizSO4-Nazminerals jadeite (Na&%zOs) and spodumene LiBOzNaJ302,20 LizCOa-NazCO3, 21 Liz(LiAlSi20a) axe found by crystal structure methN O O ~ - N ~ ~ M OLi2W04-Na2W04.22 O~,~~ The exods to be isomorphous with diopside (CaMgperimental data are indecisive or insufIicient for S ~ Z O S ) The . ~ ~symmetry is that of the monoclinic the nitrates, borates, carbonates, molybdates. of which diopside is the type, but the pyroxenes, s; in all the other cases, conspodumene structure is somewhat distorted by ibility is found a t the liquidus, the substitution L i d for CaMg, allowing the Si-0 the chlorides, bromides, and the fluoaluminates chains to pack somewhat closer together. It is forming continuous series with a minimum, to be expected, therefore, that spodumene will be while the sulfates present two series with an only partially miscible with the other pyroxenes. intermediate compound, analogous in principle to -LizSic)s3 (Fig. 2 ) . On the basis Summary s, it is probable that the iodides The phase equilibrium relations in the F. Wpmbach, Z. aswg. Ckam.. 66, 408 system, Na.$i~-Li&iOa-SiOz are described in cs Jahrb. Mineral. Geol., Beilage Bd., 48, this paper. The work was carried out by the (16) Rellner, Z.aawg. Chcm.. 157 (1917). method of quenching supplemented by thermal (17) P. Droasbach, Z. 85 (1936). (18) Thermal data by veth, J . Phrs. Chem., 4, 209 analysis. (1898); calculations b y P. dt. Cesaris, AI2i accad Lincei. 80-1, 749 There is one ternary compound, NaLiSiOs, '1911). (19) R. Nackw, Nenes Jahrb. M ~ n m a l .Geol., Beiiage Bd..14, 1 which is an end-member of the solid solution (1907) series (N&,NaLi)SiOa. It melts incongruently a t (20) H. S. van Klaoster, 2. anorg. Chcm., BB, 122 (1911). (21) W. Eitel and W Skaliks, Z. anorg. ollgmm. Chcm., lW, ZW3 847'. At the liquidus in the system, the pri(1929). c
( 2 2 ) F. Hoermann, %btd.. 177, 145 Q'328)
(23) B E Warren and J Riscor Z K r i d
80, 391 (1931)
Oct., 1939
THERMODYNAMICS OF ANILINE-NITROBENZENE MIXTURES
mary phases are (Na2,NaLi)SiOa solid solutions, Li2SiO3, NazSizO~,and LizSizOa, all three of which form solid solutions of limited extent, and the three modifications of silicon dioxide, namely, quartz, tridymite, and cristobalite. The liquidus fields meet at two ternary eutectics: one a t 697O, with (Naz,NaLi)SiOa, Li2SiOa and NazSiz05, the other a t 637') with Li~Si205,Na2Si205 and quartz as the eutectic constituents. LizSiz05melts incongruently throughout its region of existence in the system, the reaction temperature descending from 1033" in the binary system, LiSiOa-SiOz, to 641 O , the peritectic end-point in the ternary system, with Li~Si205,Li2Si03, NazSiiO5 and liquid in coexistence.
[CONTRIBUTION FROM THE
2877
The inversion temperature of quartz and tridymite has been redetermined. The temperature of 870 * loo, given by Fenner in 1913, is confirmed, the value obtained being 867 + 3 '. Refractive indices of glasses of various compositions in the system were measured. A discussion of solid solution relationships of sodium and lithium compounds in general is given, with particular reference to the theoretical aspects of the subject. Minor revisions of the phase relations in the systems, Na2SiO&3Oz and Li~Si03-SiOz, particularly with respect to the polymorphic behavior of NazSi206 and LizSi206, are presented. WASHINGTON, D.
GEOPHYSICAL LABORATORY, CARNEGIE
C.
RECEIVED AUGUST2, 1939
INSTITUTION OF WASHINGTON]
Pressure-Volume-Temperature Relations in Solutions. III. Some Thermodynamic Properties of Mixtures of Aniline and Nitrobenzene BY R. E. GIBSONAND 0. H. LOEFFLER A few months ago we noted' that the absorption of light by mixtures of aniline and nitrobenzene was strongly influenced by pressure changes. In investigating this effect we thought it desirable to examine thoroughly the volume changes which take place when these two liquids are mixed in different proportions a t different pressures and temperatures. Furthermore, we have noticed in the pure liquids2 that, whereas the internal pressure ( b E / b V ) T varies with temperature a t constant volume, another quantity, P A , computed from the P-V-T data is independent of temperature a t constant volume and is expressible as a'/ v". We have tentatively identified PA as the attractive internal pressure of the liquid. In this paper we shall give data from which the volumes of aniline-nitrobenzene mixtures may be determined a t all concentrations, between 25 and 85" and between 1 and 1000 bars, and we shall examine the behavior of PA in these solutions.
Experimental Results The solutions were all made up independently from weighed amounts of samples of aniline and nitrobenzene purified to a degree which already has been described.2 The specific volumes of (1) R. E. Gibson and 0. H. Loeffler. J . Phgs. Chum., 48, 207
(1939). (2) R . R . Gibson and 0 T I . T,oeffler, THISJ O V R N A I . , 81, 2.515 (1939).
these solutions were determined a t 25.00 O in 55-ml. U-tube pycnometers, and the thermal expansions were measured8 a t 10' intervals from 25 to 85" in a weight dilatometer of 18 ml. capacity made of vitreous silica. The volumes of the solutions were expressed as functions of the temperature by empirical equations of the form =
v65
f a(t
- 55) + b(t - 55)'
+ c(t - 55)'
(1)
We already have published for the pure components such equations whose coefficientswe may call u:, ui, b:, bi, etc. From them we computed4 ATv! and ATv; at each temperature, using 55" as the base temperature, and thence the quantity ATv - (xIAT~: x ~ A ~ = o ~8 ) ~ . This function measures the departures of the thermal expansions
+
(3) For details of the procedure in these measurements the reader is referred to the second paper of this series (ref. 2). (4) The symbols used here are as follows. The subscripts 1 and 2 refer to nitrobenzene and aniline in solution, respectively, the superscript 0 implies the pure components, and symbols without subscripts refer to the solutions. The weight and mole fractions are given by x and X ,respectively, Rz = Xz/Xi, and Y and V mean the specific and molal volumes. For a solution V = v/(xi/Mi x * / M z ) , M being the molecular weight. 9 is the apparent molal volume of a component. Av and AV are the volume changes on mixing per gram and per mole of solution. All volumes are given in milliliters. The pressure in kilobars is given by P, the temperature by 1 (centigrade scale) or T (absolute), and the total energy by E . AT and A p denote the finite changes with temperature and pressure, respectively, of the quantities to which they are prefixed. k = - a p v / v p = ~ . zI 1/(1 R%fl/Vp), z.2 = ( 1 - 21) are thevolume frartioas of the coniponentr
+
-
+
