chemical thermodynamic properties apicid internal rotation about the

SCOTT,MESSERLY, TODD, GUIHRIE, AND MCCTJLLOUGH

1320

light of the known properties of the compound C10*.'2 (12) (11 J. F. White, M. C. Taylor and G. P. Vincent, Ind. Enu. Chsm., 84, 782 (1942); (b) M. I. Sherman and J. D. H. Strickland, Anal. Chem., 27, 1778 (1955); (c) H. Dodgen and H. Taube. J. Am. Chem. Soc., 71 2501 (1949).

Vol. 65

Acknowledgments.-One of the authors (M.K.) wishes to express his appreciation to Dr. R. H. Goeckermann and N. A. Bonner of the Radiochemistry Group of the Lawrence Radiation Laboratory, Llvermore, California, for the opportunity to investigate this problem further during the Summer of 1060.

I-IEXAMETHYLDISILOXAKE : CHEMICAL THERMODYNAMIC PROPERTIES APICID INTERNAL ROTATION ABOUT THE SILOXANE LIXICAGE1

Downloaded by UNIV OF NEBRASKA-LINCOLN on August 27, 2015 | http://pubs.acs.org Publication Date: August 1, 1961 | doi: 10.1021/j100826a010

RP D. W. SCOTT,J. F. MESSERLY, S. S. TODD, G. B. GUTHRIE,I. A. HOSSENLOPP, R. T. MOORE, ANN OSBORN,W. T. BERGAND J. P. MCCULLOUGH Corifrihi~iori,Vo. 106 from the Thermodynamics Laboratorit, Bartlesville Petroleum Research Center, Bureuu of ;Ifines,

U.S. Department of the Interior, Bartlesville, Oklahoma Received January 10,1961

Thermodynamic, spectroscopic, and molecular structure information was used to show that internal rotation about an Si-0 bond in hexamethyldisiloxane is free or nearly so. Thermodynamic functions for hexamethyldisiloxane in the ideal gas state (0 to 1500°K.) were calculated by methods of statistical mechanics. Experimental studies provided the following information: values of heat capacity for the solid (11°K. to the triple point), the liquid (triple point to 375°K) and the vapor (363 to 500'K.); the triple point temperature; the heat of fusion; thermodynamic functions for the solid and liquid (0 to 375°K.); heat of vaporization (332 to 374'K.i; parameters of the equation of state; and vapor pressure (309 to 412' K.). Thermodynamic functions also were calculated for the related substance tetramethytsilane.

Studies of hexamethyldisiloxane, (CH3)3SiOSi(CH3)3,were made as part of t'hermodynamic research on organic derivatives of the lighter elements. In these shdies, the height of the barrier restricting internal rotation about a siloxane bond was determined for the first time. The experimental work consisted of low temperature calorimetry, vapor flow calorimetry and comparative ebulliometry ; detailed result's are given in the Experimental section. For convenience, the results needed for discussing the barriers to internal rotation and the thermodynamic functions are collected in Table I. TABLE I OBSERVED

T.OK.

332.31 351.50 373.67

AND

CALCULATED THERMODYNAMIC PROPERTIES OF HEXAXETHYLDISILOXANE

Entropy, S o , ral. d e g . 3 mole-' Calcd. Obsd.

134.25 137.73 141.6G

134.25 137.73 141.07

T,OK.

363.20 393.20 429.20 465.20 500.20

Heat capacity, Cpo. cal. deg.-' mole-' Obsd. Calcd.

64.45 07.68 71.40 74.95 78.16

64.44

67.67 71.41 74.93 78.17

Molecular Structure, Vibrational Assignment and Barriers to Internal Rotation Certain differences in physical properties between methyl silicones and hydrocarbons have heen attributed to freer internal rotation about the Si-0 bond than about the C-C bond. These differences include the lower temperature coefficient of viscosit)y of silicone oils and the wider tempera(1) This research waa supported by the United Statea Air Force and the Advanced Research Projects Agency of the U. 5. Department of Defense through the Air Force Office of Scientific Research of the Air Research and Development Command under Contract No. CSO 59-9. A R P A Order No. 24-59, Task 3. Reproduction in whole or in part is permitted for any purpose of the United States Government.

ture range of elasticity of silicone rubbers. However, the height of the barrier restricting internal rotation about an Si-0 bond never had been determined before the present study of hexamethyldisiloxane was undertaken. Determining the barrier height required accounting for all degrees of freedom of the molecule. To that end, moments and reduced moments of inertia were calculated from molecular structure data, the fundamental vibrational frequencies were obtained from the molecular spectra or estimated by normal coordinate calculations, and the barrier height for the methyl rotations was transferred from the related substance tetramethylsilane. Moments of Inertia.-The product of principal moments of inertia and the reduced moments of inertia for internal rotation were calculated by the general methods of Kilpatrick and Pitzer.2 The values used for bond distances and angles, based in part on electron dif@ction results, are Si-0, 1.63 K . ; Si-C, 1.88 A., C-H, 1.09 K . ; Si-043, 135'; all other angles tetrahedral. For this structure, the product of principal moments of g.3 cm.6; the average inertia is 7.126 X "effective" reduced moments of inertia (taken so their product over all internal rotations equals [D], the determinant of the internal rotational kinetic energy matrix) are 5.252 X g. for methyl rotations and 86.12 X g. cm.2 for rotations about the Si-0 bonds. The symmetry number is 2 for over-all rotation and 3 for each of the internal rotations. Spectra and Normal Coordinate Treatment.The molecular spectra of hexamethyldisiloxane (2) J. E. Kilpatrick and K. S. Pitzer, J . Chem. Phys., 17, 10G4 (1949). (3) K. Yamasaki, A. Kotera, M. Yokoi and Y. Ueda, {bid., 18, 1414 (1950).

August, 1'361

1321

C H E M I C A L PROPERTIES O F H E X A M E T H Y L D I S I L O X A N E


have been studrLedre~eatedly.~Raman and infrared frequencies that appear well established are summarized in 'Table 11.

secular equations were solved first for a trial set of force constants by Southwestern Computing Service, Inc., Tulsa, Okla., and later, after the set of force constants had been revised on the basis of TABLE I1 the first results, a t the Los Alamos Scientific VIBRATIONAL SPECTRA OF HEXAMETHYLDISILOXANE, CM.-' Laboratory, courtesy of Dr. F. H. &use. Raman, llqulda Infrared, Ilqrlld Interpretation For the final calculations, the bond stretching force constants were: Si-C stretching, 2.723 ; ? 69-86 Si-0 stretching, 4.288; interaction between a pair Skeletal bending, bl & a2 179 d of Si-C or Si-0 stretching coordinates with a silicon Skeletal bending, a1 190 p Skeletal bending, b2 & al atom in common, 0.131; and interaction between 219 d the two Si-0 stretching coordinates, -0.423; Skeletal bending, bl 253 d all X lo6 dynes cm.-'. The angle bending force 331 Skeletal bending, b2 329 d constants were: Si-C-H bending, 0.442; M i - C S i 4 stretching, a1 5% 520 p and Mi-0 bending, 0.625; S i - M i bending, 6 L9 Si-C stretching] bl 610 0.500; and interaction between a pair of CSi-C S i 4 stretching, a1 662 p or C S i - 0 bending coordinates with a Si-C or Si-0 687 Si-C stretching, bl, a2 & bl 685 d bond in common, 0.076; all X lo-" ergs radian-2. 751 d 756 Si-c stretching, a1 835 p 7 832 CHI rocking, bl All other interactions were neglected. Calculated and observed frequencies are com850 CHa rocking, a1, bl, ap & bs 848 d pared in Table 111. The assignment of observed CI.18 rocking, a, 890 d skeletal stretching and bending frequencies is 2 X 520 = 1040 Ai 1044 p straightforward. The interpretation of the ob1060 S i 4 stretching, bl served CH, rocking frequencies is reasonable but 32!jG Sym, CHI bending, ai,b, &2 & b, 1254 ignores the polarization results for the 835 and 14 10 Unsym. CH1 bending, 1405 d 890 cm.-' Raman bands. Apparently, the six 1438 al, bl, a2 & b2 lowest CHs rocking frequencies, calculated 809Region from 1450 to 2800 ern.-' omitted 812 em.-', have not been observed. The low 2900 C-H stretching 2900 intensity of these last frequencies is understandable 29GO 2960 al, bl, as & b2 if the vibrational modes are related to the CHd Region above 3000 cm.-l omitted rocking modes of tetramethylsilane of species f a p, polarinad; d , depolanzed. (inactive) and e (permitted but not observed in the Xormal coor&nate calculations were made as an Raman effect). Further discussion of the fundaaid in interpreting the observed spectra and in mental vibrations is deferred until the barriers to estimating ullobserved frequencies. The Wilson internal rotation have been considered. GI: matrix method5 was used, and the higher C-H Barrier Heights.-The barrier height for thc stretching and CH3 bending frequencies were methyl rotations was taken to be 1600 cal. mole-' factored out. 'The degrees of the secular equations as in tetramethylsilane. This value for tetrawere reduced thereby from 20, 19, 14 and 14 to methylsilane was obtained from a re-examination 10, 9, 6 and 6. An approximate potential func- of the calorimetric and spectral data for that comtion mas assumed, and force constants were trans- pound as discussed in the Appendix. Values from ferred from tet tamethylsilane for the CH3 rocking, microwave spectroscopy for methylsilane (1700 =t Si-C stretching and C-Si-C bending coordinates 100 cal. mole-')6 and dimethylsilane (1665 f 10 and estimated for the other coordinates. The cal. mole-')' are in atisfactory agreement with the

}{ }{

C %I,( ULATED ---1a-------

CH3 rocking

AND

Calcd.

iz [ 812

T4nr F: 111 OBSERVED FREQCENC~ES o r ~IEXAME'I'BYLDISII.OXANE, Cu -1 Obsd.

890 850

...

Si-C stretching Si-0 stretching C-Si--C bending

563 ,( 311 '1 220

520

...

C-Si-0 bending

169

(219) 1!10

Si-.() Lii bendiijg

82

...

___.___

-b-, Calcd.

846 824 812 704 617 1089 350 333 1 G5

7

Obsd.

(850) 834

... 686 619 1060 253 170

-?-*.-

Calcd.

Obsd.

846 812 809 i04

(850)

y b ? - - Calcd. Obsd.

(850)

(686)

847 812 809 706

(6%)

235

...

315

330

10.1-

(170)

214

219

. . ...

... ...

_.

M. J. Hunter, J . Am. Chem. Roc., 69, 803 (19.47); I. Simon and II. 0. McMahon, J . Chtm. Phys., 20, 905 (19.52); H. Murata and M. Kumada. ibid., 21, 945 (1953); R. Sh. Malkovich and V. A . Kolesova, Zhur. Fiz. Khim., 28, 926 (1954); R. Ulbrich, Z. 2Vaturforsch.. 9b, 380 (1954); C. C. Cerato. J. L. Lauer and H. C. Beachell, J. Chem. Phys., 2 2 , 1 (1954); Ya. M. Slobodin, Ya. E. Shrnulyakovskil and K. A. Rahedsinsksya, Doklody Akad. Nauk S.S.S.R., 105, 95s (1955); A. P. Kreshkov, Yu. Ya. Mik(4) N. Wright and

hailenko and G. F. Yakimovich. Zhur. Fis. Khim., 28, 537 (1954): C. A. Frensel, D. W. Scott, and J. P. McCullough, Bureau of Mines Report of Investigations No. 5658 (1960). ( 5 ) E. Bright Wilson, Jr.. J. C. Deoius and P. C. Cross, "Molecular Vibrations," McGraw-Hill Book Co., Inc.. New York, N. Y.,1955. (6) R. W. Kilb and L. Pierce, J . Chem. Phys.. 27, 108 (1957). (7) L. Pierce, ibid., 31, 547 (1959).


1322

SCOTT,MOORE,OSBORN,BERGA X D MCCULLOUGH

thermodynamic value for tetramethylsilane. This observation shows that the barrier for a methyl group attached to silicon is relatively independent of the molecular environment, as was assumed when the barrier height for tetramethylsilane was transferred to hexamethyldisiloxane. Comparisons with the calorimetric data were made to investigate internal rotation about the Si-0 bonds. When calculated values were used for all unobserved vibrational frequencies, and internal rotation about the Si-0 bonds was assumed to be free, both the entropy and the heat capacity were calculated to be greater than the observed values. Agreement between calculated and observed entropy could be obtained by selection of a suitable barrier height for the siloxane rotation, but only a t the expense of a greater discrepancy between the calculated and observed heat capacity. On the other hand, agreement for both entropy and heat capacity could be obtained by suitable adjustment of unobserved vibrational frequencies if the siloxane rotation was assumed to be free. These observations are evidence for relatively free siloxane rotation. For calculating thermodynamic functions, the siloxane rotations were taken to be completely free, although the thermodynamic evidence does not rule out a modest barrier height of a few hundred cal. mole-'. The explanation of the unusual physical properties of methylsilicones as arising from relatively free internal rotation about Si-0 bonds is confirmed by the thermodynamic results. Another explanation that has been advanced, weaker intermolecular forces between methylsilicone molecules, can be rejected because the gas imperfection data for hexamethyldisiloxane reported in the Experimental section indicate normal intermolecular forces. Fundamental Vibrational Frequencies.-The unobserved skeletal bending frequencies of species al, calculated 82 and 341 cm.-', depend upon the Si-0-Si bending force constant, for which only an order-of-magnitude estimate was used in the normal coordinate treatment. Therefore these frequencies may be adjusted to conform with the thermodynamic data. Also, the average value for the six unobserved CH3 rocking frequencies, calculated 809-812 cm.-l, may be adjusted within reason. The values actually taken for theseeeight unobserved frequencies, selected to fit the thermodynamic data, are 102,360 and 820(6) cm.-'. The complete set of fundamental vibrational frequencies used for calculating thermodynamic functions is listed in Table IV. Average or conventional values were taken for the C-H stretching and CH3 bending frequencies, which are not all resolved in the observed spectra and make unimportant contributions to the thermodynamic functions except a t high temperatures. Thermodynamic Functions The molecular structure parameters described in the preceding paragraphs were used to compute the values of thermodynamic functions listed in Table V. Empirical anharmonicity contributions,8 with ( 8 ) J. P. McCullough, H. L. Finke, W. N. Hubbard, W. D. Good, R. E. Pennington, J. B. hlesserly and C. Waddington, J . A m . Chem. Soc.. '76, 2661 (1954).

T'ol. 65

TABLE IV FUNDAMENTAL VIBRATIONALFREQUENCIES OF HEXAMETIIYLDISILOXANE,~ CM,-1 C-H stretching CHZunsym. bending CHa sym. bending

at

bi

a2

b2

2950 (5) 1420 (3) 1255 2) 890

2950 (5) 1420 (3) 1255(2) (850) 834 18201 686 619 1060 253 12351 179

2950 ( 3 ) 1420 (3) 1255 (850) {820] {820) (686)

2950 (4) 1420 (3) 1255 (850) {820} (820) (686)

CHI rocking Si-C stretching

{ :2

Si-0 stretching

520

{(E;

[235] 330 C-Si-C bending C-Si-0 bending 190 (179) 219 Si-0-Si bending I1021 a ( ), observed frequency used more than once; [ 1, from normal coordinate calculations; { ) , selected t o fit calorimetric data.

v = 1130 cm.-l and 2 = 0.52 cal. deg.-l mole-', were included to give better agreement with the experimental values of heat capacity a t the higher temperatures. The contributions of anharmonicity are only 0.001 and 0.005 cal. deg.-' mole-' in So and Cpo a t 298.15OK. but increase to 0.48 and 0.88 cal. deg.-' mole-' a t 1500OK. Calculated values of So and Cpo are compared with the observed values in Table I. Agreement well within the accuracy uncertainty of the experimental data was obtained over the entire temperature range of the experiments. TABLE V THEMOLALTHERMODYNAMIC FUNCTIONS OF HEXAMETHYLDISILOXANE I N THE I D E A L GASSTATE4 IF0

T. OK.

-

I H O

-

Ei"o)/T,

*-o)/T,

deg.-1 cal.

deg.-l cal.

Hoekcal. H a,

So, deg.-l oal.

Cp', deg.-l cal.

0 0 0 0 0 0 -89 61 33.40 273.15 9.122 123.00 54.02 -92.61 35.24 298.15 10.51 127.85 57.00 -92.83 35.38 300 10 61 128.20 57.22 16.90 146.23 68.40 400 -103.98 42.26 500 -114.10 48.49 24.24 162.58 78.16 600 -123.44 54.14 32.48 177.58 86.48 700 -132.18 59.29 41.50 191.47 93.67 800 -140.41 63.99 51.19 204.40 99.95 -148 19 68.31 900 61.48 216.49 105.48 -155.58 72.28 72.28 227.86 110.33 1000 -162.64 75.94 83 54 238.58 114.59 1100 -169.39 79.33 95.19 248.72 118.33 1200 1300 .-175.86 82.46 107 20 258.32 121.60 85.37 119.52 267.44 124.47 -182.07 1400 1500 -188.05 88.06 132.10 276.12 126.99 a To retain internal consistency, some of the values are given to one more decimal place than is justified by the absolute accuracy.

Experimental The basic experimental techniques are described in published accounts of apparatus and methods for low tem(9) The contributions of vibration and anharmonicity were oomputed by the Bureau of Mines Electronic Computer Service, Pittsburgh, Pa. The contributions of restricted internal rotation were computed b y Denver Electronia Computing Service, Inc., by twoway curvilinear interpolation in the tables of K. s. Pitser and W. D. Gwinn, 3. Chem. Phys., 10, 428 (1942).

Alugust,l!I61

CHEMICAL PROPERTIES OF HEXAMETHPLDISILOXAYE

perature calorimetry,1° vapor flow calorimetry,ll and comparative ebulliometry.lz The reported values are based on a molecular weight of 162.384 g. mole-' (1951 International Atomic Weight@), the 1951 values of fundamental physical constants1*and the relations: 0" = 273.15OIi.l6 and 1 cal. = 4.184 joules (exact1.y). Measurements of temperature were made with platinum resistance thermometers calibrated in terms of th.e International Temperature Scale,16 between 90 and 500CK.,aiid the provisional scaleI7 of t,he National Bureau of Standards, between 11 and 90°K. All elect,rical and mass n-teasurements \yere referred to standard devices calibrated a t the National Bureau of Standards. Material.--Dow Corning Corp. purified grade (9991,) hexamethyldisiloxme was used as starting material for highly efficient, fractional distillation and silica gel percolation, xhich was done by C. J . Thompson and H. J . Coleman of this Center. The combined best fractions used for low temperature calorimetry and comparative ebulliometry had a purit,:: of 99.996 mole To,determined by calorimetric studies of melting point as a function of fract,ion melted. Combined wcond-best fraciions were used for vapor flow calorimetry Heat Capacity in the Solid and Liquid States.-The observed values of heat capacity, Cantd, are list'ed in Table VI. Above 3OoK., the accuracy uncertainty is estimated to be no greater than 0.2%. The heat capacity curves ( Csatd cs. T ) are normal for both crystals and liquid. The observed d u e s fcr the liquid may be represented within 0.0570 betneeri the triple point and 375°K. by the empirical equation Cs,td(liq) = 65.834t - 8.2399 X 10-'T 5.1874 X 10-4T1 - 4.8933 X 10-72'3, cal. deg.-l mole-' (1) Heat of Fusion, Triple Point Temperature, Cryoscopic Constants and Purity of Sample.-Values of the heat of fusion, AHnz, from three determinations were 2848.4, 2848.9 and 2850.7 cal. mole-'. The accepted value is 2849.4 =t 1.5 cal. mole-'. The results of a study of the melting temperature, Tobs(f, as a function of the fraction of total sample melted, F are listed in Table VII, Also listed in Table VI1 are the values obtained for the triple point temperature, T T . ~ .the , mole fraction of impurity in the sample, N", and the cryoscopic constants18 A = A H m / R!&-.P.~ and B = ~ / T T . P -ACm/2Am, . calculated from the observed values of TT.P., AHm. and A C m (the heat capacity of the liquid less thnt of the solid, 6.42 cal. dcg.-l mole-'). Thermodynamic Properties in the Solid and Liquid States.-Values 01: thermodynamic functions for the condensed phases were computed from the calorimetric data for selected temperatures between 10 and 3750K0. The results are given in Table VIII. The values at 10 h. were computed from a Debye function for 7.5 degrees of freedom with e = 101.73'; these parameters were evaluated from the heat capacity data beta-een 12 and 21°K. Corrections for the effwts of premelting have been applied to the "smoothed" data i~xordedin Table YIII.


I

+

TABLE VI THERIOLAL HEATCAPACITY O F HEXAMETH YLDISILOXANE IN CAL.DEG.-~ T,O K , a Csstd Csatd 133.15 43.804 Crystalss 139.00 45.206 11.57 1.641 140.52 45.540 12.77 2.087 117.36 47.175 14.35 2.757 154.52 48.824 3.402 15.78 161.96 50.494 16.64 3.795 169.68 52.206 17.64 4.256 175.62 53.495 4.709 18.62 177.18 53.867 19.68 5.195 182.47 55.040 20.47 5.566 183.22 55.230 21.48 6.029 22.42 184.49 55.476 6.457 189.83 56.704 6.913 23.40 190.63 56.866 7.493 24.68 197.86 58.551 8.008 25.86 Liyiiid 8.518 27.03 208.94 66.776 9.291 28.90 216.95 67.398 9.632 29.76 222.34 67.766 10.643 32.21 226.86 68.133 12.159 35.98 231.28 68.465 13.665 39 85 236.66 68.910 15.117 43 67 246.83 69.769 16.773 47.91 257.35 70.651 17.694 50 22 267.76 71.601 18.649 52.65 278.04 72.504 20.028 56.17 288.19 73.475 22.501 62.45 298.23 74.430 24.832 68.74 299 85 74.568 27.241 75.51 309.61 75.526 29. E10 82.26 320.33 76.597 31.840 88.97 330.90 77.641 32.812 92.22 341.35 78.684 33.847 95.71 351.66 79.736 35.237 100.54 361.84 80.771 37.433 108.39 370.89 81.667 39.436 115.86 41.277 122.99 0 T is the mean temperature of each heat capacity measurement. Csatdis the heat capacity of the condensed phase Values of Csatd for crystals are not at saturation pressure. corrected for the effect of premelting. T,'K.a

t

__ ~.

(10) H. hl Huffm.an, Chem. Rev., 40, 1 (1947); H. A I . Huffman, S. S. Todd and G. :D. Oliver. J . Am. Chem. Soc., 71, 584 (1949); D. IT. Scott, D. R. Douslin, RI. E. Gross, G. D. Oliver and H. hf. Huffman, ibid., 74, 833 (1952). (11) G. W:iddingtan, S. S. Todd and H. RI. Huffman, ibid., 69, 22 (1947); 6. €'. McCullough, D. W. Scott, R. E. Pennington, I. A. Hossenlopp and G. Waddington, ibid., 76, 4791 (1954). (12) G. WiLddingtcn, J. W. Knoalton. D. W. Scott, G.D. Oliver, S. S. Todd, W. N. Hubbard, J. C. Smith and H. hl. Huffman, ibid., 71, 797 (1949). (13) E. Wichers, itid.. 74, 2447 (1952). (14) F. D. Rossini F. T. Gucker, Jr., H. L. Johnston, L. Pauling and G. W. Vinal, ibzd., 74, 2699 (1952). (15) Some of the xesults originally were computed with constants and temperatures in terms of the relation 0' = 273.16'K. Only results affected signilicantly by the new definition of the absolute temperature scale [E[. F. Stimson, A m . J . Phys., Z3, 614 (195511 were recalculated. Therefore, numerical inconsistencies, much smaller than the accuracy uncertainty, may be noted in some of the reported data. (16) H. F. Stimson, J . Research Natl. Bur. Standards, 42, 209 (1949). (17) H. J. Hoge a n d F. G. Brickwedde, {bid., 22, 351 (1939). 1. Streiff a n d T. D. Rogsini, ibid.. 36, 35.5 (lQ45h

1323

TABLE VI1

HEXAMETHYLDISILOXANE: MELTINGPOIKT SUMMARY TT.P.

= 204.93

0.00004;

A =

1cIelt ed , %

?C 0.05OK.; 0.03414 deg.?;

Nz*= B

AF(TT.P.

Tobad, 1/F

-

Tabsd)

=

= 0.00375 deg.-l Tgrapb,

01c.a

OK.

204.9165 204.9165 8.805 ,9217 3.845 ,9217 .9236 .9236 1.986 ,9242 1.432 ,9241 ,9245 ,9245 89.28 1.120 100.00 1.000 ,9246 0 204.9257 Pure Temperatures read from a straight, line through a plot TabsdV S . 1/F. 11.36 26.01 50.34 69,83

Q

Of

Vapor Pressure .-Observed values of vapor pressure are listed in Table IX. The condensation temperature of the sample was 0.004' lower than the ebullition temperature at 1 atm. pressure. The Antoine and Cox equations selected to represent. the results are log p = 6.77651 - 1203.556/(t 208.427) (2)

+

SCOTT,GUTHRIE,HOSSENLOPP, MOOORE AND MCCULLOUGE-I

1324

TABLE VI11 THE MOLAL THERMODYNAMIC PROPERTIES METHYLDISILOXANE I N

OF

HEXA-

SOLID AND LIQUID STA?’E90 Csatd.


cal. des. - 1

10 12 14 16 18 20 25 30 35 40 45 50 80 70

80 90 100 110 120 130 140 150 160 170 180 190 200 204.93

0.091 .156 .246 .359 .496 .655 1.133 1.‘705 2.345 3.034 3.762 4.517 6.091 7.728 9.407 11.113 12.834 14.563 16.288 18.009 19.721 21.421 23.108 24.779 26.433 28.077 29.702 30.50

Crystals 0.274 2.74 .470 5.64 .716 10.02 1.008 16.12 1.336 24.04 1.690 33.80 2.651 66.27 3.659 109.77 4.672 163.53 5.682 227.27 6.680 300.6 7.674 383.7 9.657 579.4 11.626 813.8 13.555 1084.4 15.441 1389.7 17.260 1726.0 19.007 2090.8 20.691 2482.9 22.313 2900.7 23.879 3343 25.393 3809 26.863 4298 28.294 4810 29.689 5344 31.05 5900 32.39 6478 33 04 6771

0.365 .626 .962 1.367 1.832 2.345 3.784 5.364 7.017 8.716 10.442 12.191 15.748 19.354 22.962 26.554 30.09 33.57 36.98 40.32 43.60 46.81 49.97 53.07 56.12 59.13 62.09 63.54

1.125 1.792 2.603 3.499 4.421 5.342 7.632 9.726 11.768 13.717 15.631 17.602 21.540 25.283 28.825 32.15 35.08 33.87 40.51 43.03 45.43 47.79 50.06 52.26 54.47 56.72 58.96 60.03

Liquid 204.93 30.50 46.94 9620 77.44 66.45 210 31.65 47.42 9958 79.07 66.89 220 33.88 48.32 10631 82.20 67.60 230 36.04 49.18 11311 85.22 68.37 240 38.16 49.99 11998 88.15 69.19 250 40.21 50.78 12694 90.99 70.03 260 42.23 51.53 13399 93.76 70.89 270 44.18 52.27 14112 96.45 71.79 273.15 44.79 52.49 14339 97.28 72.07 280 46.09 52.98 14835 99.07 72.69 290 47.96 53.68 15567 101.64 73.66 298.15 49.46 54.23 16170 103.69 74.49 300 49.80 54.36 16308 104.16 74.58 3 10 51.59 55.03 17058 106.62 75.57 320 53.35 55.68 17819 109.03 76.56 330 55.07 56.33 18590 111.40 77.55 340 56.76 56.97 19370 113.73 78.55 350 58.42 57.60 20161 116.02 79.57 360 60.05 58.23 20961 118 28 80.59 370 61.66 58.84 21772 120.50 81.58 375 62.45 59.15 22181 121 60 82 06 0 The values tabulated are the free energy function, heat content function, heat content, entropy and heat capacity of the condensed phases a t saturation pressure. and log (p/760) = A( 1 - 373.669/2‘) log A = 0.891266 9.2338 X 10-4T 9.0805 X lO-’T* (3) I n these equations, p is in mm., t is in “C., and T is in OK. The observed and calculated vapor pressure for both the

-

+

Vol. 65

Antoine and Cox equations are compared in Table IX. The normal boilingpoint is 100.52” (373.67”K.).

TABLE IX OF HEXAMETHYLDISILOXANC VAPORPRESSURE Boiling point, OC. Ref. Hexamethylcompd.0 disiloxane

p(obsd.),b mm.

p(obad.) Antoine eq. 2

19.061 21.720 24.388 27.068 29.757 32,460

36:206 38.984 41,777 44.585 47.407 50,241

71.87 81.64 92.52 104.63 118.06 132.95

-0.02

(35‘174)} 60.000 65 70 75

53.099

149.41

-

- p(ca1cd.) cox eq. 3

.01 .Ol .02

-0.02 .01 .02 .04 .04 .05

.03

.00

.oo

+ .01

+

58.837 187.57 .04 - .02 64.625 233.72 .01 .00 70.469 289.13 .01 .00 80 76.363 355.22 .09 .06 85 82.320 433.56 .07 .03 90 88.330 525.86 .06 .02 95 94.399 633.99 .OL .04 100 100.520 760.00 .05 .02 105 106.696 906.06 .07 .02 110 112.932 1074.6 .1 .o 115 119.222 1268.0 .u .o 120 125.567 1489.1 .2 .o 125 131.971 1740.8 .1 .2 130 138.417 2026.0 .7 .1 0 The reference compound from 71.87 to 149.41 mm. was pure benzene; that from 149.41 to 2026.0 mm. was pure water. b From vapor pressure data for benzene [F. D. Rossini, K. S. Pitzer, R. L. Arnett, R. M. Braun and G. C. Pirnentel, “Selected Values of Physical and Thermodynamic Properties of Hydrocarbons and Related Compounds,” Carnegie Press, Pittsburgh, Pa., 1953, Table 21k] and for water IN. S. Osborne, H. F. Stimson and D. C. Ginnings, J . Research Natl. Bureau Standards, 23, 261 (1939)]. Heat of Vaporization, Vapor Heat Capacity and Effects of Gas Imperfection.-The experimental values of heat of vaporization and vapor heat capacity are given in Tables X and X I . The estimated accuracy uncertainty of the values of LWVand C,” are 0.1 and 0.2oj,, respectively. The heat of vaporization may be represented by the empirical equation AHV = 11794 2.5429T - 2.4172 X 10-2Tz, cal. mole-’ (332 - 374’K.) (4) The effects of gas imperfection were correlated by the procedure described in an earlier paper.l0 The empirical equation for B , the second virial coefficient in the equation ofstate,PV = RT(1 f B / V ) , i s B = 560 - 282.1 exp(800/1’) cc. mole-’ (332 - 500°K.) (5) “Observed” values of B and -T(dZB/dT2) = 1im@Cp/

+

+

-

-

+ -

+

+

-

P+O

dP)T and those calculated from eq. 5 are compared in Tables X and XI. The observed gas imperfection is

TABLE X THEMOLALHEATOF VAPORIZATION AN) SECOND VIRIAL COEFFICIENT OF HEXAMETHYLDISILOXANE T,

OK.

p, atm.

AHv, cal.

B(obsd.), CC.

B(calcd.), cc.b

-2493 -2572 0.250 8280f4a 332.31 0 500 7914f2n -2191 -2187 351.50 1.000 -1856 -1840 7469 f 2a 373.67 0 Maximum deviation from the mean of three or more determinations. b Calculated from eq. 5. (19) J. P. MoCullough, H. L. Finke, J. F. Messerly, R. E. Pennington, I. A. Hossenlopp and G. Waddington, J . Am. Chem. Sac., 17, 6119 (1966).

August , 1961

CHEMICAL PROPERTIES OF HEXAMETHYLDISILOXANE

1325

TABLE XI THEMOLALVAPORHEATCAPACITY OF HEXAMETHYLDISILOXANE IN CAL.DEG.-' T,

363.20

OK.

Cp (1 .OOO atrn.) Cp (0.500 atm.) Cp (0.333 atm.) Cp (0.250 atm.) CPO

- T(dtB/dT')"


a

65.273 64.846 64.45

393.20 68.942 68.288 68.074

429.20 72.215 71.793

465.20 75.467 75 * 202

500.20 78.556 78.354

67.68

71.40

74 * 95

78.16

0.76 .74

0.48 .52

0.38

obsd. 1.53 1.14 calcd.' 1.58 1.09 Units: cal. dog.-' mole-' atm.-I Calculated from eq. 5.

.30

roughly that expected for a substance of the molecular TABLE XI11 size and shape of hexamethyldisiloxane and with normal F U N D A M E N T A L irIBRATIONAL FREQUENCIES intermolecular forces. OF TETRAMETHYLSILANE,~ Cm.-1 The heat of vaporization at 298.15"K. was calculated by extrapolation with eq. 4 (8.89 kcal. mole-'), by use of the a1 e fl fz Clapeyron equation with eq. 3 and 5 (8.87 kcal. mole+) C-H stretching (2967) (2967) 2967 and by use of a lthermodynamic network with the thermoC-H stretching 2900 (2900) dynamic functions of Table V (8.90 kcal. mole-'). From CHs unsym. bending 1418 (1418) 1430 the last value, selected as the most reliable, and eq. 5, the standard heat of vaporization was calculated, L W V ~ = ~ S SCH, . ~sym. ~ bending 1250 1263 8.92 kcal. ~ n o l e - ~ CHs rocking [813] [SO91 862 Entropy in the Ideal Gaseous State.-The entropy in S i 4 stretching 593 694 the ideal gaseous state at 1 atm. pressure was calculated Skeletal bending 199 245 as shown in Table XII. (), observed frequency used more than once; [I, from TABLE XI1 normal coordinate calculations.

THEMOLALENTROPY OF HEXAMETHYLDISILOXANE IN THE IDEAL GASEOUS STATE IN CAL. DEG. A Cox vapor pressure equation was fitted to the 351.50 373.67 data of Aston and co-workers because it is more 332.31 T, OK. (licl)" AHvIT S(idea1) S(real)b

Ssstd

-

R In Pc

111.95 24.92 0.14 -2.76

116.37 22.51 0.23 -1.38

121.31 19.99 0.36 0.00

- - -

S"(obsd.) f 0 25d 134.25 137.73 141.66 By interpolation in Table VIII. The entropy in the ideal gas state less that in the real gas state, calculated from eq. 5. Entropy of compression, calculated from eq. 3. d Estimated accuracy uncertainty. (1

Appendix Barrier to Internal Rotation and Thermodynamic Functions of Tetramethylsilane.-Calorimetric20 and spectralZ1 data for tetramethylsilane were reexamined to determine the height of the barrier restricting the methyl rotations. The set of fundamental vibrational frequencies taken for thermodynamic calculations is in Table XIII. The two unobserved CHa rocking frequencies were calculated in the course of normal coordinate calculations to obtain force constants for transfer to hexamethyldisiloxane. The values obtained (813 and 809 cm.-l) do not differ grossly from the values calculated by KovalevZ2 (855 and 829 cm.-') and by Shimizu and Murata** (830 and 825 cm.-'). For Si-C distance 1.888 A., C-H distance 1.10 A.24 and all angles tetrahedral, the product of principal moments of inertia is 2.012 x 10-lla g.3 cm.O,and the reduced moment of inertia for internal rotation is 5.292 X g. cm.* (20) J. 0.Aaton, R. M. Kennedy and G. H. Measerly, J . Am. Chem. SOC.,63, 2348 (1941). (21) D. H. Ranh, B. D. Saksena and E. R. Shull, Disc. Faraday Soc., 9, 187 (1950). (22) I. F Kovalev, Optzka i Speclroekopia, 6, 387 (Eng. ed.) (1959). (23) K. Shimizu and H. Murata, J . Molecular Spsclroacopu, 6 , 44 (1960). (24) W. F. Sheeban, Jr.. and V. Schomaker. J . Am. Chem. Soc., 74, 3956 (1fl52).

reliable for obtaining derivatives than the type of four-constant equation originally used by those workers. With this Cox equation

-

log (p/760) = A( 1 299.80/T) log A = 0.819737 9.1552 X 10-4T 1.2078 X 10"T2

-

+

and an estimated value for the second virial coefficient (-2.7 1.)) the heat of vaporization a t 227'K., calculated by use of the Clapeyron equation, is 6750 cal. mole-l. This value is 2.5% lower than the one originally reported by Aston, et al. TABLE XIV THEMOLALTHERMODYNAMIC FUNCTIONS OF TETRAMETHYLSILANE IN THE IDEAL GASSTATE; RIGIDROTATOR, HARMONIC OSCILLATOR, INDEPENDENT INTERNAL ROTATOR APPROXIMATION" T. OK.

IFo;/T, CBI.

deg.-1

#O-

o)/T, cal. deg.-l

H"

-

H'a,

kcal.

SO, cal. dej.-'

CPO,

cal.

dsy.

-1

273.15 -62.00 20.87 5.701 82.87 32.61 298.15 -63.87 21.02 6.537 85.79 34.39 300 -64.00 22.00 6.600 86.00 34.52 400 -70.89 25.99 10.40 96.88 41.30 500 -77.10 29.67 14.83 106.76 47.32 600 -82.81 33.05 19.83 115.86 52.53 700 -88.15 36.16 25.32 124.31 57.06 800 -93.16 39.03 31.23 132.19 61.04 900 -97.91 41.69 37.52 139.59 64.57 1000 -102.42 44.14 44.14 146.56 67.67 1100 -106.73 46.41 51.05 153.14 70.40 1200 -110.86 48.51 58.21 159.37 72.79 1300 -114.82 50.46 65.60 165.28 74.88 1400 -118.62 52.28 73.19 170.90 76.71 1500 -122.28 53.96 80.95 176.24 78.32 To retain internal consistency, some of the values in this table are given to one more decimal place thaD is justified by the absolute accuracy.

W. N.HUBBARD, F.R. FROW AND GUYWADDIYGTON

1326

Vol. G5

Addition of the revised value of entropy of vaporization to the calorimetric value of entropy for the liquid gives for the gas, So227 = 77.00 cal. deg.-' mole-l, This value of S0227and the original value based on a calorimetric determination of of So298.18, the heat of vaporization, were used in calculating the barrier height. At both temperatures, a value close to 1600 cal. mole-' is obtained; this is the value transferred to hexamethyldisiloxane, as already discussed. T , OK. #"(free int. rot.), cal. deg.-' mole-' S"(obsd.), cal. deg.-' mole-'


Sr - S, cal. deg.-' mole-' Barrier height, cal. mole-'

The various molecular structure parameters for tetramethylsilane were used in computing the values of thermodynamic functions in Table XIV. The calculated values of entropy, = 77.12 and SO298.16 = 85.79 cal. deg.-' mole-', may be compared with the experimental values given in the previous paragraph. As effects of aiiharmonicity were not taken into account and no vapor heat capacity data were available for checking, the thermodynamic functions for tetrsmethylsilane in Table XIV are less reliable than those for hexa227 298.16 methyldisiloxane in Table 1'. The values in 80.73 88.31 Table XIV differ from those published by Shimizu 77.00 85.79 -~ and Murata23 mainly in the use of a value of barrier 3.73 2.52 height based on a reinterpretation of l s t o i i and co1651 1598 workers' data.

THE HEATS OF COMBUSTION AND FORMATIOX OF PYRIDINE AKD HIPPURIC ACID1 BY W. N.HUBBARD, F. R. FROW A N D GUYWADDINGTOX~ Contribution AVO. 101 f r o m the Thermodynamics Laboratory, Petroleum Research Center, Bureau of Mznes, U . S. Department of the Interior, Bartlesville, Oklahoma Recezved January 10, 1962

The heats of combustion of pyridine and hippuric acid Fere determined by precision oxygen-bomb calorimetry, and these values, in kcal. mole-', are reported for the standard heat of formation, AHfo29~.ls,from graphite, and gaseous hydrogen, oxygen and nitrogen: pyridine(liq.), 23.89; hippuric acid(c), - 145.54. Recommendations are made for methods of determining the amount of reaction in combustion calorimetry. The suitability of hippuric acid as a reference substance for the combustion calorimetry of organic nitrogen compounds is discussed.

Heat of combustion measurements are part of a continuing study by the Bureau of Mines of the thermodynamic properties of organic nitrogen compounds important in petroleum technology. This paper gives the results of heat of combustion measurements on pyridine and presents related thermochemical data that were used in an earlier comprehensive report on the chemical thermodynamic properties of this substance.3 Also given are the results of measurements on hippuric acid, which has been proposed as a reference substance II H I

H-C

I

,Y

0 I'

C'--C-

H I X-CH2-C

0

//

H H

for combustion calorimetry of organic nitrogen compounds.4 (1) This investigation was performed as part of American Petroleum

Institute Research Project 52 on "Nitrogen Constituents of Petroleum," which is conducted a t the University of Kansas in Lawrence, Kansas, and a t the Bureau of Mines Petroleum Research Centers in Laramie, Wyoming, and Bartlesville, Oklahoma. (2) Requests for reprints should be addressed to the Thermodynamics Laboratory, Bartlesville Petroleum Research Center, Bureau of Mines, U. s. Department of the Interior, Bartlesville, Oklahoma. (3) J. P. McCullough, D. R. Douslin, J. F. Messerly. I. A. Hossen. lopp, T. C. Kincheloe and Guy Waddington, J. Am. Chem. Sac., 79, 1289 (1957). (4) H. M. Huffman, i b d , 60, 1171 (1938).

Experimental Materials.-The sample of pyridine used for this study has been describeds; it was dried in the liquid phase by calcium hydride and handled in a vacuum distillation system at all times. The sample of hippuric acid was prepared by recrystallizing commercial material three times from distilled water. Part of theorecrystallized material was dried in a vacuum oven a t 80 , and part was dried in air a t 105 to 110'. An attempt to purify part of the recrystallized material further by zone refining failed because the sample decomposed a t a temperature only slightly higher than the melting point. Pellets of dried hippuric acid were stored over Pz05. Units of Measurements and Auxiliary Quantities.--Xll data reported are based on the 1951 International Atomic Weight@ and fundamental constants6b and the definitions: 0°C. = 273.15" K.; 1 cal. = 4.184 (exactly) joules. The laboratory standard weights had been calibrated a t the National Bureau of Standards. For use in reducing weights in air to in vacuo, in converting the energy of the actual bomb process to the isothermal process, and in reducing to standard states,7 the values tabulated below, all for 25O, were used for density, p , specific heat, cp, and ( B E / ~ P )ofT the substances. e,, cal, deg.-L g.-'

g. % - I

Pyridine Hippuric acid

0.978 1.371

Calorimetry.-The

0.401 ,286

bomb, Ta-1 ,E

(aE/ap)~,

cal. atm.-l g.-1

- 0.0076

-

,0027

and rotating-bomb

( 5 ) R. V. Helm, W. J. Lanuin, G. L. Cook and J. S. Ball, J . Phys. Chem., 68, 858 (1958). (6) (a) E. Wichers, J . Am. Chem. Soc.. 74, 2147 (1952). (b) F. D. Rofisini, F. T. Gucker, Jr., H. L. Johnston, L. Pauling and G. W. Vinal, ibid., 74, 2699 (1952). (7) W. N. Hubbard, D. W. Scott and G. Waddington, "Experimental Thermochemistry," F. D. Rossini, Editor, Intersoience Publishers, Inc., New York, N. Y . , 1956, Chapter 5 , PD. 75-128. (8) W. N. Hubbard, J. W. Knowlton and H. M. Huffman, J . Phys. Chem., 68, 396 (1954).

chemical thermodynamic properties apicid internal rotation about the

Recommend Documents