(4) Comings, E. W., Clapp, J. T., Taylor, J. F., Ind. E n & Chem. 40, 1076-82 (1948). (5) Davis, L., J e t Propulsion Lab., Pasadena, Calif., Rept. 3-22, 1950. (6) Ibid., 3-17, 1952. (7) Douglas, T. B., Furukawa, G. T., Coskey, R. E., Ball, A. F., J. Research Natl. Bur. Standards. 53, 139-53 (1954). (8) Fro’ssling, N., Gerlands Beitr. Geophys. 51, 167-73 (1937). (9) Zbid., 52, 170-216 (1938). (IO) Groot, S. R. de, “Thermodynamics of Irreversible P r o c e s s e s , ” Interscience, New York, 1952. (11) Hirschfelder, J. O., Curtiss, C. F., Bird, R. B., “Molecular Theory of G a s e s and Liquids,” Wiley, New York, 1954. (12) Hsu, N. T., Reamer, H. H., Sage, B. H., Am. Doc. Inst., Washington 25, D. C., Document 4219 (1954). (13) Hsu, N. T., Sage, B. H., Ibid., Document 5311 (1957). (14) Hsu, N. T., Sato, K., Sage, B. €I. Ind. , Eng. Chem. 46, 870-76 (1954). (15) Hughes, R. R., Gilliland, E. R , Chem. Eng. Progr. 48, 497-504 (1952). (16) Ingebo, R. D., Natl. Advisory Comm Aeronaut., Tech. Note 2368 (1951). (17) Zbid.. Tech. Note 2850 (1953). (18) International Critical Tables, vol. 5, McGraw-Hill, New York, 1929. (19) Kirkwood, J. G., Crawford, B., Jr., J . Phys. Chem. 56, 1048-51 (1952). (20) Kramers, H., Physica 12, 61-80 (1946). (21) Langmuir, I., Phys. Revs. 12, 368-70 (1918). (22) Lewis, G. N., J. Am. Chem. SOC.30, 668-83 (1908). (23) Lewis, G. N., Proc. Am. Acad. Arts Sci. 37, 49-69 (1901). (24) Linton, W. H., Jr., Sherwood, T. K., Chem. Eng. Progr. 46, 258-64 (1950). (25) McAdams, W. H., “Heat Transmission,” 3rd e&, McGraw-Hill, New York, 1954. (26) Maisel, D. S., Sherwood, T. K., Chem. Eng. Progr. 46, 131-88 (19502
Zbid. 46, 172-5 (1950). Meyers, C. H., Bur. Standards J. Research 9, 807-13 (1932). Morse, H. W., Proc. Am. Acad. Arts Sci. 45, 361-7 (1910). Page, F., Jr., Corcoran, W. R,Schlinger, W. G., Sage, B. H., Am. Doc. I n s t , Washington 25, D. C., Document 3293 (1952). (31) Powell, R. W., Trans. Inst. Chem. Engrs. (London) 18, 36-50 (1940). (32) R a w , W. E., Marshall, W. R, Jr., Chem. Eng. Progr. 48, 141-6, 173-80 (1952). (33) Rossini, F. D., r o t h e n , “Selected Values of P h y s i c a l and Thermodynamic Properties of Hydrocarbons and Related Compounds,” Camegie P r e s s , Pittsburgh, 1953. (34) Sage, B. H., Chem. Eng. Progr. 49, Letter to Editor (July 1953). (35) Sato, K., Sage, B. H., “Thermal Transfer in Turbulent Gas Streama Effect of Turbulence on Macroscopic Transport from Spheres,” Paper 57-A-20, Heat Transfer Division, Annual Meeting, New York, N. Y., December 1957, Am. SOC. Mech. Engrs. (36) Schlinger, W. G., Reamer, H. H., Sage, B. R, Lacey, W. N., “Report of Progress-Fundamental Research on Occurrence and Recovery of Petroleum, 1952-1953,” Am. Petroleum Inst., pp. 70-106. (37) Schlinger, W. G., Sage, B. H., Am. Doc. I n s t , Washington 25, D. C., Document 4221 (1954). (38)Schubauer, G. B., Natl. Advisory Comm. Aeronaut., Rept. 524 (1935). (27) (28) (29) (30)
Received for review July 1, 1957. Accepted December 20, 1957. Material supplementary to this article h a s been deposited as Document No. 5600 with the AD1 Auxiliary Publications Project, Photoduplication Service, Library of Congress, Washington 25, D. C A copy may be secured by citing the document number and by remitting $2.50 for photoprints or $1.75 for 35-mm. microfilm. Advance payment is required. Make checks or money orders payable to Chief, Photoduplication Service, Library of Congress.
Specific Heats of Aviation Hydraulic Fluids ROGER S. PORTER and JULIAN F. JOHNSON California Research Corp., Richmond, Calif. I t is useful to know the specific h e a t s of hydraulic fluids over wide temperature ranges. F o r t h e severe conditions under which such fluids perform in modern supersonic a i r craft, specific heat is quantitatively considered in heat transfer calculations in aircraft design. It is instrumental in determining rates of heating or cooling under unsteady s t a t e conditions. T h e choice of a hydraulic fluid with a given specific heat w i l l thus influence the transient thermal response characteristics of t h e hydraulic system and will partially define i t s thermal performance. Empirical correlations have proved very satisfactory for predicting the specific h e a t s of the common pure liquids and of certain liquid mixtures (9, 15, 20). Unfortunately, estimating specific heats of common hydraulic fluids by this approach is very difficult. First, fluids are often compounded from substances of greatly differing molecular t y p e T h i s results in large heats of mixing and makes unreliable any additive rule for predicting specific heats of fluids from their pure components. There is a dearth of heat capacity data on liquids which are chemically related to the components in many hydraulic fluids. Moreover, the few available values are not in good agreement and do not generally cover a large temperature range. Therefore, it i s necessary to evaluate the specific h e a t s of the hydraulic fluids experimentally. HYDRAULIC FLUIDS
T h e three hydraulic fluids investigated a r e not pure compounds but mixtures compounded for unique and desirable properties (5). Aircraft hydraulic fluid MILO.5606 is a petroleum b a s e fluid which is recommended for u s e below
272
INDUSTRIAL AND ENGINEERING CHEMISTRY
7OoC. T o date, over 25,000,000 gallons of t h i s fluid have been placed in operation. A typical composition of t h i s fluid is shown in Table I. Oronite high temperature hydraulic fluid 8200 was d e veloped to provide a fluid with excellent physical properties
Table I. Composition of Fluid MIL-0-5606 Wt. % Highly treated light g a s o i l fraction Highly treated heavy g a s oil fraction Poly alkylmethacrylate Oxidation inhibitors Red dye
60-80
15-30 4-8
0. M . 5 Trace
for use in aircraft at elevated temperatures. T h e fluid is composed predominantly of a specific alkoxydisiloxane. It contains a silicone thickener which a c t s a s a viscosity index improver. Oronite high temperature hydraulic fluid 8515 h a s essentially the same composition a s fluid 8200, except that it contains 15% by weight di(2-ethylhexyl) sebacate. T h i s gives fluid 8515 a greater compatibility with rubber in hydraulic systems over the recommended operating range of -54 t o 204 OC. METHOD
A differential heating method w a s chosen for measuring specific heats. T h i s choice w a s based on t h e nature of the VOL. 3, NO. 2
hydraulic fluids, the wide temperature range over which results were desired, and the e a s e with which measurements can be made tiy t h i s method. Differential heating techniques have received relatively l i t t l e attention in the literature, although a simple and successful method based on t h i s a p proach h a s been developed by Spear (16). T h e equipment used i n t h i s research was very similar to that of Spear. However, modifications were made; and it is felt that the additional contributions of t h i s work have made t h e technique more workable and accurate. In a differential heating method, results are obtained by comparing t h e rates of heating for a t e s t liquid with that for a fluid of known specific heat. T h e conditions for heating the t v o fluids are made a s nearly identical as possible. T h e liquids a r e contained in light metal v e s s e l s which are
of water before they are placed in the bath. A s soon as the t u b e s are in the bath, a stethoscope i s used to check the activity of the magnetic mixers. A prominent pinging sound is evidence of good mixer action. Voltages are recorded on each fluid a t alternate minutes to t h e nearest microvolt. Temperature data are taken in t h i s way from a t least 15OC. below to 15'C. above each temperature a t which the specific heat i s desired. Sample tubes are reweighed at the end of the measurements. Thermocouple voltages are converted to temperatures and plotted against elapsed times. T h e best line is drawn through the data, and the times required for the c e l l s to p a s s through equal arbitrary temperature intervals a r e cdculated. T h i s temperature interval i s from 12' to 28°C.. depending on the rate of change of temperature with time.
ELECTRIC STIRRER
suspended in air jackets. They a r e then allowed to heat up in similar fashion towards the constant temperature of a surrounding bath. in such experiments, specific heat of the test liquid is calculated from the heat capacity of the reference fluid, the weights of the sample fluids, the thermal capacities of t h e two cells, and the times required for the cells to go through the same temperature interval. T h e value obtained i s the mean specific heat of the fluid over a short tqnperature range. Figure 1 shows a diagram of the apparatus. T h e sample tubes are made of brass 1/16 inch in thickness, 7/8 inch in diameter, and 6% inches in h g t h . The t u b e s and their c a p s are nickel plated. It was necessary to build recesses 1%inches in diameter and inch in depth in the b a s e of the liquid bath container directly opposite the b a s e s of the brass cylinders. T h i s permits the a i r d r i v e n magnetic stirrers (Precision Scientific Co.) to b e placed close enough to drive the magnets within the sample tubes. T h e force field of t h e magnetic stirrers i s concentrated by using shaped bars of Armco iron. T h e liquid bath c o n s i s t s of Dow Corning 710 fluid stirred by a hallow-bladed stirrer. T h e bath is controlled by a Thermotrol temperature regulator (Hdlikainen Instruments). Regulation is 0.01 OC. from 25" to 22OoC. Magnetic stirrer speed is measured by u s e of a Strobotac. T h e thermocouples a r e Chromel-Alumel. Their voltages are measured using a L e e d s & Northrup White double potentiometer. T h e liquid bath temperature is s e t at l e a s t 25°C. above the highest t e s t temperature. T h e sample t u b e s are filled with 20 ml. of sample, weighed, and inserted in the constant temperature bath. For mom temperature measurements, the sample t u b e s are precooled to slightly above the dew point
The t e s t temperature is at t h e midpoint of t h e range. Usually, tan intervals are evaluated at each t e s t temperature. Several t e s t temperatures may be calculated from o n e experimental run. Specific h e a t s are Calculated from Equations 1 and 2 (4, 16).
k, + CP, W,
.
1958
f,
+R,=
k, + CP, W,
+R,
(1)
f.
where
k,, k2 =thermal capacities of cell 1 and cell 2 Cp,, Cp, = h e a t capacities of liquids in cell 1 and cell 2
W,,W, = weights of liquids in cell 1 and cell 2 R,, R,
= evaporation equivalence factors for cell
1
and cell 2 f , , f a = t i m e s required for cell 1 and cell 2 to go through the same temperature range Corrections for losses due to evaporation were necessary only for diphenyl ether above 180'C. T h e correction calculation was based on the following assumptions: that the heat loss is proportional t o the product of the vapor pressure and the elapsed time, that h e a t s of evaporation remain constant over the temperature interval, and that Trouton's rule is applicable. From t h e assumptions, R may b e calculated as:
where
AHv = h e a t of vaporization per unit weight CHEMICAL AND ENGINEERING DATA SERIES
273
was used. T h e diphenyl ether w a s a 25% center distillation cut from Eastman Organic Chemicals No. 104. I t s freezing point w a s 26.77OC. Literature values (18) a r e 26.85' and 26.90'C. T h e heptane w a s also a center distillation cut of Phillips 99 mole % with a freezing point of -90.66OC., identical with t h e reported value (1).
14 -
-
.d
>w
13RESULTS
a
3 U
e
12-
50
IO0
I
200
I50
TEMPERATURE,^. Figure 2. Thermal capacities of sample tubes
W T =weight loss of sample during experiment P, = final vapor pressure a t time tf P b = vapor pressure a t time tb P a = vapor pressure a t time t , t = tb - t , AT = temperature range over which specific heat i s calculated T h e thermal capacities of t h e c e l l s , k, and k,, can be calculated from t h e weights of t h e materials used in constructing t h e sample c e l l s (#, 16). However, increased accuracy can be obtained by calibration with liquids of known specific heats. Calibrations also compensate for variations in t h e way in which heat is transferred from t h e constant temperature bath t o t h e thermocouples within t h e cells. To reduce t h e amount of experimental calibration required, it w a s assumed that t h e thermal capacity of cell 1 was correct a s calculated from i t s construction. Using liquids of known heat capacities in both cells, t h e thermal capacity of cell 2 is then calculated from Equation 1. Results obtained i n t h i s way a r e shown in Figure 2. Over-all accuracy of t h e calibration was confirmed by observed cow sistent specific heat data for different weights of t e s t sample and accurate results on fluids of known specific heat, examples of which a r e shown in T a b l e 11. Water and heptane a r e used as standards in t h e region from 2 5 O to about 30°C. below their respective boiling points. Diphenyl ether is used from 80' t o 22OoC. Values for specific h e a t s of t h e standards a r e known t o about kO.2% (6, 11). Distilled water boiled to remove occluded g a s e s
Table
60°C.
II. Sample Accuracy T e s t s a t
Specific Heat
Reference Fluid
Test Fluid
Calcd.
Correct
Error
n-Heptane n-Heptane Diphenyl ether
n-Heptane Water Water
0.57 3 0.997 0.997
0.569
0.7 0.3 -0.3
7 0
LOO0 LOO0
Table
+ -
T h e specific h e a t s of three hydraulic fluids have been measured by a differential heating method at atmospheric pressure from 25 to 220 C. Results for fluid MIL-0-5606 and Oronite high temperature hydraulic fluid 8515 a r e best interpreted by linear equations relating specific heat and temperature. Data for Oronite high temperature hydraulic fluid 8200 show a marked increase i n t h e change of specific heat with increasing temperature, and in consequence results a r e best presented in quadratic form. Results for fluid MIL-0-5606 are based on three experimental temperature-time curves. Specific heat values for fluids 8200 and 8515 represent a minimum of s i x experimental runs for e a c h fluid with runs made in duplicate over each of three temperature intervals. T h e results of t h i s research a r e shown in equation form in T a b l e 111. Deviations l i s t e d in specific heat values a r e t h e standard errors of estimate. T h e s e represent t h e 67% confidence limits for experimental values. T h e experimental results plotted i n Figures 3, 4, and 5 show change of spe-
w
8.55-I
' S O
