3.F. NAYLOR
2120 [CONTRIBUTION FROM THE PACIFIC
EXPERIMENT STATION, BUREAUOF
VOl. 67 MINES, UNITED STATES
DEPARTMENT OF THE
INTERIOR]
High-Temperature Heat Contents of,Na,TiO,, Na,Ti,O, and Na2Ti,0,1 BY B. F. NAY LOR^ Investigation of the thermodynamic properties carbon dioxide. Analyses for titanium and carof metallurgically important titanium compounds bon dioxide were carried out on each compound, has been part of the recent program of the Pacific On the basis of the titanium analyses the Na2Experiment Station. Smelting of titaniferous T i 4 , NarTipOs and NaaTi,O7 samples were iron ores with sodium carbonate and carbon judged 98.4, 98.9 and 98.6% pure, respectively. forms sodium titanates as part of the reaction The carbon dioxide analyses, reported as Narproducts. The thermochemical data essential for COa, gave 0.5,0.3and 0.3%) respectively. studying the thermodynamics of this reaction do Results not exist in the literature. As partial fulfillment of Lhis need, the present paper reports heat conExperimental results are presented in Table I tents above 298.1GOK. of three sodium titanates, and Fig. 1. The column labeled T , OK., lists from room temperature to about 1,600°K,, from which specific heats, heats of fusion, and enTABLE I tropies above 298.16’K. have been calculated. HEATCONTENTS ABOVE 298.16 OK. (CAL./MOLB)
-
-
NatTiOa Method and Materials (mol. wt. NarTiSOa NstTinOr 141.894) (mol. wt. 221.794) (mol. wt. 301.694) The measurements were made by the “drop” HT HT Hi9r.a HT T,OK. Hmii method in an apparatus previously de~cribed.~ T,OK. H n t s . 1 6 T,O K . The calorimeter was calibrated with electrical 38a. 7 2,925 362.8 3,078 374.3 4,725 513.4 7,220 474.3 8,530 energy, measured in international joules, and the 478.6 11,400 531.3 7,890 573.4 13,640 results were converted to the conventional calorie 624.4 21,300 by the relation, 1 cal. = 4.1833 int. joules (NBS).4 655.0 8,940 705.7 20,650 738.8 29,260 589.7 10,390 814.0 26,580 Therinocouple calibrations, made frequently in 843.5 36,700 638.0 12,120 924.9 32,510 the manner described by Southard, were based on 962.7 45,310 765.6 16,9DO 996.2 36,480 1079.2 53,920 the melting points of gold and palladiuni. 948.6 24,410 1034.4 38,680 During the measurenients the samples were en1169.4 60,660 1198.9 63,040 closed in platinum-rhodium alloy capsules, the 1056.9 28,970 1039 6 39,040 necks of which were pinched off and sealed with 1096.3 30,770 1109.1 43,45O(P) 1260.7 68,24O(p) 62,360 1197.7 50,06O(P) 1431 111,080 platinum. The heat content of the empty cap- 1414 68,090 1216.1 53,02O(P) 1446 sules had been determined separately; a srnall 1536 113,540 72,320 1361 83,840 1484 121,750 correction for the platinum used in sealing was 1584 1381 1524 127,150 86,380 applied. 1514 1606 134,870 95,310 At about 800” in the case of the NatTi03, and 1579 1681 142,000 99,fj50 at somewhat higher temperatures for the other titanates, small pin-holes began to appear in the capsule walls. During measurements in the the temperature of the sample immediately before liquid range some of the sample extruded through dropping into the calorimeter and HT - €1298.16, these holes. Simultaneously, a small increase in the heat liberated per gram molecular weight in weight of the capsule was noted due to reaction of cooling from T to 298.16OK. The values thought the sample with carbon dioxide and moisture. A to involve premelting have been designated close check was maintained and weight correc- “(#)”. The listing order is not necessarily the tions were made. New capsules and fresh order in which measurements were made; howsamples were substituted when the increase in ever, determinations a t lower temperatures generally preceded those a t higher Lemperatures. sample weight amounted to more than 0.2%. All molecular weights accord with the 1941 InThe titanates were prepared in this Laboratory by treating stoichiometric weights of NazCOa, ternational Atomic Weights. Sample weights prepared from reagent-grade NaHCOs, with J. T . were corrected to vacuum using the experimenBaker TiOz, which analyzed 98.6% pure. The tally determined densities: NazTiOa, 3.19; Nalreaction mixtures were heated at 900 to l,lOOOfor TiaOs, 3.40; NazTiaOT, 3.44 g./cc. No correction several hours with constant pumping to remove of the heat contents was made for the minor impurities present. (1) Published b y permission of the Director, Bureau of Mines, The melting points are indicated by dotted U . S. Department of the Interior. N o t copyrighted. 12) Formerly chemist, Pacific Experiment Station, Bureau of lines in Fig. 1. Insufficient measurements were Mine¶. made near the melting points to allow independent (3) J. C . Southard, T m s JOURNAL, 63, 3142 (1941). determinations of the melting point temperatures (4) E . F . hlucller arid F. D . Rossini. Am. J. Pky.+s. 12, 1--7 The temperatures used in this paper are those (lY44)
-
I
HIGH-TEMPERATURE HEATCONTENTS OR SODIUM TITANATES
Dec., 1945
2121
given by Washburn and Bunting.' By extrapolating the heat-content data to their reported temperatures, the heats of fusion per mole were calculated to be 16,810 cal. for N%TiO, at 1,303'K., 26,230 cal. for NarTiSOa a t 1,258'K., and 37,100 cal. for NaZTi-07 at 1,401'K. The metatitanate has a small transition a t 560'K., the heat &ect of which was computed to be 400 cal. per mole. However, the transition temperature was calculated from the experimental data and may be in error possibly by as much as 20'. Heat-content equations, fitting the experimental data closely, were derived. They are given below, followed by the appropriate temperature range and, in the case of the solids, by the mean percentage deviation from the experimental results. The heat contents thought influenced by premelting were not used in computing the mean deviation. No mean deviation is given for the liquid range because it was possible to determine only a few experimental points in each instance and this precluded a significant calculation. (The equations were based upon the following smooth, molal heat content values: Na2TiOs(a) 3,300 at 400'K. and 6,750 at 500'K.; NazTiOa(B) 10,750 at 600'K., 18,320 a t 800'K. 200 000 1000 1400 1800 and 30,950 at 1,100'K.; NaaTiOs(l) 61,720 at T,OK. 1,400'K. and 71,100 a t 1,600'K.; NazTizO&) 4,880 a t 400'K., 20,350 a t 700'K. and 36,750 at Fig. 1.-High temperature heat contents: A, NalTiOs; 1,000'K.; Na2TizO~(l)87,400 at 1,400'K. and B, NarTbOs;C, NaaT406. 101,100 at 1,600'K.; NazT&O7(s)6,360 a t 400'K., 26,550 a t 700'K. and 63,000 a t 1,200'K.; Naz- 100' intervals and the results are listed in Table TiaOl(l) 124,890 a t 1,500'K. and 143,720 at 11. 1,700'K.). TABLE I1 NapTiO:(cu): HT - H ~ B . = , s 26.18T 0.01036Ta- 8429 HEAT CONTENTS U D ENTROPIES ABOVE 298.16O K . (298-560'K.; 0.1%)
-
+ 25.95T+ 0.00850P - 7880 (560-1303'K. : 0.1%)
- H208.18 NazTiOs(1): HT - Hw.~,,= 46.9T-'3940 (1303-16OO0K~) NasTiOs(B): HT
NanTizo6(s):HT
-
Hs8.16
4G0'000
NazTitOb(1): HT-
T
= 49.32T
- 16,560 (298-1258'K.;
0.3%)
H 2 0 a . 1 6
+
+
T
-
HW.16
=
94.15T
-
1G.335 (1401-1700°K.)
The corresponding specific heat equations are
+ +
NanTiOa(a): C, = 25.18 0.02072T NazTiOp(B): Cp = 25.95 0.01700T NazTiOa(1): CD = 46.9 460,000 NazTizO(s): CD = 49.32 0.00706T - Tp NazTiaOs(1): cp 68.5 564,000 = 63.46 0.01064T Na*Ti2G(s):*CD P NazTi307(l): cp= 94.15
+
p
+
Haw.rc
RT
+ 0.00353P +
68.5T -'8500 (1258-1600°K.) NapTi:O,(s): HT - HZP(.I~ = 63.46T 030532P 5fi4'000- 21,285(298-1401'K.; 0.15%) NaZTbQ(1): HT
-NasTiOp-
-
Using the above equations, heat contents and entropies above 298.16'K. were computed a t (6) E. W. Washburn and E. N. Bunting. Bur. SIardardr J . R .. #*arch, I t , 319 (1984).
400 600 660
560 600 700 800 900 1000 1100 1200 1258 1258 1300 1303 1303 1400 1401 1401 1500 1600 1700
ST
-
SZo8.11
(cal./ (cat./ dcg./ mole) mole) 3,300 9.61 6,750 17.20 8,920(0) 21 .30 9,320(@) 22.01 10,760 24.48 14,450 30.18 18,320 35.34 22,360 40.10 26,570 44.53 30,950 48.71 35.500 52.66 40,220 40,360(8) 67,170(1) 61.720
56.44 56.55 69.45 72.82
66,410 71,100
76.08 79.08
cNaaTisOlST
Rzw.u
HT
(cp1.l mole) 4,880 9,900
15,070 20,350 25,730 31,200 36,750 42,380 48,090 51,440(s) 77,67Q(1) 80,550 87,400 94,280 101,100
-
Sw.18
-NaaTub ST
HT-
5'asa.a
-
(ca1.l de&/ mole) 14.06 25.26
Haw.ic (ca1.l mole) 6,360 12,900
(deg. mole( 18.32 32.91
34.67 42.81 49.99 56.43 62.28 67.65 72.64 75.34 96. I9 98.44
19,650 36,550 33,590 40,i60 48,060 55,470 63.000
45.20 55.84 65.24 73. 6Y 81.37 88.44 95.02
70,640
101.10
103.62
78,390 78,470(s) 115,570(1) 108.24 124,890 112.66 134,300 143,720
(cal./
106.E4 106.90 133.38 139.81 146.89 151.39
Discussion
The heat contents above 298.16'K. of many substances can be calculated with fair accuracy by adding together the known heat contentr of
K. J. PALMER AND MERLEB. HARTZOG
2122
Vol. 67
their constituents. Parks and Kellef have shown this to be true for calcium and magnesium silicates, and later Kelley7 based his calculation of the entropy and free energy of formation of sodium oxide upon the same principle. Values for the heat content above 298.16’K. of Na20, obtained by subtracting the heat contents of one, two and three moles of Ti02*from the heat contents of N a ~ T i 0 ~NaTi20s ,~ and Na2TiB07,respectively, have been listed in Table 111. X similar extraction for Na20, also given in Table 111, was computed using the heat con-
tents of NazSiOa and NaaSi20~.’O As Si02 exists in several €orms, the heat content of SOz, used in this instance, was obtained by taking the difference between the heat contents of Nafiiz06 and Na2Si03. The mean of the calculated heat contents is shown in a separate column in Table 111. The following algebraic equations were derived and use of them in thermodynamic calculations is suggested pending actual measurements of Na20. Sa.0: IIT - H298.16 = l5.9T f 0.0027P - 4980
TABLE 111 CALCULATED HEATCONTENTS ABOVE 298.16”K. OF NanO (CAL./MOLE)
(6) G. S. Parks and K. K. Kelley, J . Phys. Chcnt., 30, 1175 (1926). 61,471 (1939). (7) K.K.Kelley, THISJOURNAL,
Summary High-temperature heat contents above 298.16’ K. of NazTiOa, NazTiz06 and Na2Ti307 were determined to about 1,600’K. The heat of fusion of each compound and the heat of transition of NazTiOa at 560’K. were calculated from these data. Heat-content and entropy increments above 298.16’K. have been tabulated a t 100’ intervals and algebraic heat content and specific equations fitting the experimental data were derived. Estimated heat contents above 298.16’K. of Nan0 a t 100’ intervals to 1,100’K. and corresponding heat-content and specific-heat equations were suggested for use in thermodynamic calculations.
( 8 ) Unpublished heat content data for TiOz. (9) Heat contents used for NarTiOi above 560’K. were obtained by extrapolation of the a-curve.
BERKELEY, CALIF.
400 500 600
700 800
900 1000 1100
1,760 3,650 5,615 7,605 9,745 12,050 14,510 17,125
1,800 3,700 5,600 7,460 9,380 11,380 13,430 15,530
1,740 3,600 5,445 7,215 9,065 11,030 13,080 15,195
1,750 3,560
5,320 7,220 9,150 11,060 13,080 15,110
1,760 3,630 5,495 7,375 9,335 11,380 13,520 15,740
N a 2 0 : C, = 15.9
+ 0.0054T
(298-1lOOOK.)
(10) B.F.Naylor, THISJOURNAL, 67, 466 (1945).
RECEIVED AUGUST9, 1945
1 CONTRIBUTION FROM WESTERN REGIONAL,RESEARCH LABORATORY1a]
An X-Ray Diffraction Investigation of Sodium Pectate BY K. J. PALMER AND MERLEB, HARTZOG
The occurrence of long chains in pectin was first suggested by Srnolenski.lb In 1930 Meyes and Mark2 proposed a structure for pectin consisting of a long straight chain of galacturonide units in the pyranose form linked together by 1:4 glycosidic bonds. The structural formula for pectic acid (completely demethylated pectin) proposed by them is shown in Fig. 1. The existence of long chains in pectin has since been confirmed by the work of Schneider and co-workers, Van Iterson
Fig. 1.-Structural formula of pectic acid as suggested by Meyer and Mark. (la) Bureau of Agricultural and Industrial Chemistry, Agricultural Research Administration, U. S. Department of Agriculture. Not copyrighted. ( l b ) Smolenski, C.A , , 19, 41 (1925) (2) Meyer and Mark, “Der Aufbau der Hochpolymeren organis&en Naturstoffe,” Leipzig. 1930,p. 219.
and others. The results of these investigations and their bearing on the pectin problem have been well reviewed in Meyer’s recent book.s In addition to this evidence the more recent investigations of the flow birefringence,”Bb sedimentation and diffusion,5”,band the viscosity studies6 of pectin solutions are-all in agreement with the concept that pectin is a long chain molecule. It also has been shown recently7 that pectin is composed essentially of a-d-galacturonide units in the pyranose form connected by 1:4 glycosidic linkages. Another aspect of the structure pro(3) Meyer, “Natural and Synthetic High Polymers,” Vol. 1 Interscience Publishers, Inc., New York, N.Y.,1942,p . 363. (4) (a) Snellman and Saverborn, Xoll. Beihcffc, 52, 467 (1941); (b) Boehm, Arch. c w p f l . ZclLforsch, Gnucbcsiichl., 22, 520 (1938-1939). (5) ( a ) Saverborn, Kolfoid Z.,90, 41 (1940); (b) Tiselius and Inpelmann, Fdrh. Svenska Suckcrjabriksdirigcnlcrnas FBrcn. Samman(rbdcn, 1042,II.16 pp.; C.A , , 98,4465 (1844). ( 6 ) Owens, Lotzkar. hlerrill and Peterson, THIS JOURNAL, 66, 1178 (1944). (7) Morrell, Baur and Link, J. B i d . Chcm., 105,l (1934); Luckett and Smith, J . Chcm. SOC.,1108 (1940); Hirst, ibid., 70 (1942).
