[CosTRIBUTIOS F R O M
THE
?rIISERAI.S THERMODYNAMICS EXPERIMEST STATIOS, REGION11, BUREALT OF STATES DEPARTMENT O F THE I N T E R I O R ]
>IINES,
I-SITEI)
Heats of Formation of Crystalline Silicates of Strontium and Barium BY R. EARANY, E. G. KINGAND S. S. TODD RECEIVEDMARCH5 , 1957 Heats r,f formation from tlie cornpo~ientoxides a t 298.15"I.c. were obtained for strontium metasilicate, strontium cirthtrsilicate, barium ~netasil;cate,barium orthosilicate, dihariurn trisilicate antl bariuni disilicate by measuring heats of reactiriti of the silicate; and oxides with hydrofluoric acid.
This paper presents heats of formation of two crystalline strontium silicates (SrSiOs and SraSi04) and four crystalline barium silicates ( BaSi03, Ba2Si0.1, Ba2Si30sand BaSiy05). Previous papers from this Laboratory by Torgeson and Sihama' and King' contained corresponding data for some magnesium and calcium silicates. Materials and Method The silicates were prepared b!- repeated, prolonged >intering of stoichiiimetric mixtures of reagent-grade strontium or barium carbon;tte and pure quartz. 111 each instance several heats were required, with intervening grinding, ]nixing, ;inalysis and adjustment of composition, to ohtain a satisfactory product. In preparing strontium metasilicate the total heating time IKLS 216 lir. between 1000 and 1350°, of which 32 hr. were above 1200". The product gave ;in X-ray diffraction pattern agreeing virtuall!. perfectly with t h a t reported by Carlson and 1veils.3 Chemical analysis gave 63 .26c)l, strontium oxide, 36.66' silica and 0.12( iron and aluminum oxides (theoretical analysis, 63.30"; strontium oxide and 36.70'1 silica). strontium orthosilicate required four heats totaling 35 lir. a t 12ilO-130Oo, ir-hich ir-ere conducted in a platinum vessel. X-lZay diffraction of the product gave no eL-idence of uncombined oxides, but the pattern disagreed with t h a t of O'Daniel aiid Tscheischwili,4 who may have had a different crystalline variety. Ana sis gave 77.515;. strontiurn oxide, 22.47('i silica and 0.03 sodium oxide (theoretical analysis, 77.52( ;, strontium o de and 22.48%; silicaj. The stoichiometric mixture for the preparation of barium metasilicate w x s pelletized and heated nine times for a total of 247 hours between 1006 and 111O0, of which 18 Imurs were above 120Oo. Silvcrnian and co-workers5 list two crvstalline varieties of barium metasilicate. ,4ustin6 obtained the S - r a J - diffraction pattern of one variety, but t h a t Of the other has not been reported in the literature. T h e pattern for the present material showed no evidence of uncombined oxides but disagreed with t h a t of Austin, indicating t h a t i t probably is the \-arietk- labeled "BaO. SiO?(II1" by SilT-erman and cci-workers. Analysis gave i1.95( barium oxide and 28.170c \ilica, as compared with the theoreticai 71.85 and 28.15', . Barium rirthodicate TKIS prepared in a nickel crucible; the stoichiometric mixture was heated five times for a total ;it 1~l(Ji1-11~5i~0 and 16 hours a t 1150-1300°. some Littack of tlie crucible, making i t necessary to discard the surfare layer after each heat. The cleaned product gave no viiihle precirjit, of nickel glyoxime from a I-g. sample. The X-ray tliffrac n pattern agreed with t h a t reported b ~ -Austiti.6 Aiialys shoiyed 83.78'); barium oxide and 16.39' silica, as compared with the theoretical 83.62 and 1B.38'(. T h e preparation r i f dibarium trisilicate required many heats and a long period of time, even though the mixture \viis strongly pelletized. The records show totals of 11 (1) D. R. Torgeson and Th. G. Sahama, THISJ O U R N A L , 70, 215ii (1948). (21 E . G. King, ibid.,73, GI3 (1951). ( 3 ) E . T. Carlson and I,, S.\Yells, J . R e s e a r c h S a i l . Birr. S t a n d a r d s , 61, 74 (19533. (4) H . O'Daniel and I,. Tscheischn-ili, Z . K r i s t . , 104, 361 (1942). ( 5 ) A . Silverman, H. Insley, G. I\', LIorey and F. D. Rossini, Natl. Research Council Bull. 118, J u n e . 19-1'3. [ t i ) A . E. Austin, .I, A r l i C r i n m . C O G , 30, 218 ( l W i ) ,
days a t 1050°, 78 lir. a t 1200", 24 hr. a t 1250°, and 6 lir. a t 1300'. The product gave an X-ray diffraction in excellent agreement with t h a t of Austin.6 Chemical malysis showed 63.0Oc,)~barium oxide, 37.03'; silica, and 0.01' iron and aluminum oxides (theoretical atialysis, 65.98' barium oxide and 3 7 . 0 2 silica). ~~ The stoichionietric mixture for preparing barium tlisilicate was pelletized and heated for a total of 100 hr. a t l l O O o and 30 hr. a t 1300". T h e X-ra>- diffraction pattern obtained immediately after the last heat checked t h a t in the ASTM Catalog. However, the pattern changed slowly with time on standing a t room temperature, indicating the compound to be unstable. I n using this substance in the subsequentlJ- reported measurements, it was reheated to 1300" and quenched to room temperature immediately before each run. Chemical analysis gave 56.05c2 barium silica, as compared with the theoretical 56.06 and 43.94' Strontium oxide was prepared from reagent-grade strontium carbonate by first repeatedly boiling in water and filtering and then heating to 1400' until decomposition was essentially complete. Analysis gave 99.75"L strontium oxide and less than O.lc!c carbon dioxide. The substance was kept in sealed glass bulbs before use in the measurements. Barium oxide was prepared from reagent-grade barium hydroxide. The latter was dissolved in water and treated with hydrogen peroxide t o precipitate h3-drated barium peroxide. The precipitate was dried, heated in 7 " x o to 1300" to decompose it to the normal oxide, and sealxl in glass bulbs for preservation. Analysis of the product showed 0.15(',1 water, 0.02'5 silica and O.ObLi, barium carbonate as impurities. The heats of formation were determined by measuring the heats of reaction of the silicates and their constituent oxides with 20.1:; hydrofluoric acid a t 73.7'. In all instances 940.1 g. of the acid was employed. The masses of the silicates, strontium oxide and barium oxide conformed stoichiometrically in each instance with 0.7420 g. of quartz, which was adopted as a b a k . The samples were contained in gelatin capsules for dropping from room temperature (measured aiid correction made to 25') into the calorimeter. Correction for the capsules was made from separate measurements of the heat of snlution of gelatin under the same calorimetric conditions. The various compounds that were run picked up small amounts of water in handling. In each instance the amount of this water was determined by weight loss o n ignition antl correction was applied for it. I t was found expedient to mix about 0.1 g. of barium carbonate per capsule with the barium compounds to prevent clumping and therefore t o obtain complete reactirm with t h e acid. T h e heat of reaction of barium carbonate with the acid was determined as -31.5 cal./g. in separate experiments, and proper correction was applied. The heat of reaction of the barium carbonate was between (1.5 and 2 ' , of the totai heat evolved, depending upon the substance under study. The calorimetric apparatus was t h a t described by Torgeson and Sahama' with the modifications reported by King.*
Measurements and Results All values are in terms of the defined calorie ( 1 cal. = 4.1840 abs. joules). A11 weighings were reduced to vacuum, and all molecular weights accord with the 1954-55 Report on Atomic n'eights.' ( 7 ) IC. \Tichers,
T H I S JOlIRNAI.,
78, 3283 ( l % 5 l ~ ) ,
3640
Each heat of formation is the resultant of tneasurements of a series of reactions for which skeleton equations are given in Table 1. The preci7im uncertainties were calculated according to the methods of Rossini and Deming.a The experimental data are given in Table I which shows the reactions, numbers of detertninxtions, average heat values, and preciiion uncertainties. In this table, "c" denotes a crystalline substance, "sol" denotes a substance in solutio:i, and "p" denotes a precipitate. The heat of reaction 1 is from the paper of who employed the same equipment and calorinetric conditions as in the present work. Each of reactions 2 to 5 were conducte.1 in snlu tions identical in compo-ition with the final solution from reaction 1. Each of reactinns 0 to 1 1 were conducted i n fresh acid, without previous arlditions. This procedure is necessary to produce the exact material balance in the summation of the reaction heats to obtain the desired hezts of fotmation of the silicates from their conctituent oxide;. The heat of reaction 3 i; nearly 500 cal. niorc than twice that of reaction 2 , showing th:tt thern 1s ' a C??? siderable concentration effect. On the other hnnrl, the concentration effect for reactions 4 a n d 5 w a ~ negligible. Eleven deterniinations were made of the heat of reaction of barium oxide with acid c m taining the proper amount of dissolved silic2. The samples of barium oxide were varied in size to cover the composition range necessary to conform stnichionietrically tyith compounds ranginn fro*n TITSir05to BasSi04. No significant dependence of the heat of reaction on sample size was noted, and so the same molal value is used in obtaining the hent; of formation of all the barium silicates. Measurement of the heat of reaction 11 wa? difficult because of the previoucly mentioned instability of barium disilicate. Even though the sample; were heated to 1300" and quenched to rami temperature, the rewlts were uncerttin. The three bect values obtained are -9S;?20, -9S.470 and -98,330 with a mean of -OS,310 f 130 c.d. Table I1 illustrates the use of the d:tt,t in Table I for obtaining heats of formation of the silicates from their constituent oxides. L
( 8 ) I'. 13 Ronsini a n d I
19:3(l)
IY.1:. DeminK, J . IT-asiz. d c a ~ i .S:i., 2 9 , 416
.is the fin 11 solutim after cmducting rezctioris 1 and 2 cmsecutivelv i!i the same acid solution is identical in c?*n*,?sitim with the solutim after con3uctiny reictinn 6 in fre;h acid, it follows that the re-tctim? and heats m w be addecl as intlic,ited to obt2in the heit of for:n,ttim of strontiux inetasilic'tte a t 25' fro*=it; con;tituent oxide;. The heits of formation of the crystal!ine strom tiuw and bnriuw .iilicntei, obtained in this mi~nner. appear in Txhle 111. Bccauye of the before mentioned di6ficultv o f ine:isureinent, it is recoinmended t,hat the v a l w for bariuin disilicitte lie interpreted as setting :in tinper limit (algebraic) to the he'it of fw:natim, rather than as clcarl:L- tlcfinin? its value. T4nr.e 1 I I
HC.%TSOF Sub.tance
SrSi?{ SrlSi04 T3aSiq2
FORM.4TIOS
i !I
FROM
TIIC
O X I D E S (I
