Znd. Eng. Chem. Res. 1996,34,404-409
404
Determination of Critical Properties (Tc, P,)of Some Jet Fuels Jian Yu and Semih Eser* Fuel Science Program, Department of Materials Science and Engineering, 209 Academic Projects Building, The Pennsylvania State University, University Park, Pennsylvania 16802
The critical temperatures of some petroleum- and coal-derived jet fuels were measured with good reproducibility, using a rapid heating sealed tube method. A critical property estimation method was used to calculate the critical temperatures and pressures of the jet fuel samples. Five other methods were also used to estimate the critical temperatures of the fuels. While most of the estimation methods give satisfactory predictions of the critical temperatures of the petroleum-derived jet fuels, they all fail to give acceptable results for the coal-derived jet fuels. The larger differences between the measured and estimated critical temperatures for the coalderived fuels are attributed to the M e r e n t composition of these fuels, dominated by cycloalkanes. On the basis of properties of the compounds found in the coal-derived jet fuels, a new correlation is proposed to calculate the critical temperatures of the coal-derived jet fuels.
Introduction The future fuel system in an advanced aircraft will be operating under supercritical conditions because of the increased thermal management requirements (Edwards and Zabarnick, 1993). At high-temperature supercritical conditions, the fuel will decompose to form detrimental solid deposits. The mechanisms of deposit formation from jet fuels under supercritical conditions are not known. In view of the fact that future jet fuels will encounter supercritical conditions, there is a need to study specifically the effects of supercritical conditions on the deposit formation from jet fuels. It is clear that the critical properties of the jet fuels must be determined to define their sub- and supercritical regions. The critical properties of substances can be either measured experimentally or estimated using some other related properties. Various experimental methods have been developed to determine the critical properties of pure compounds and complex mixtures such as petroleum fractions (Roess, 1936; Kobe and Lynn, 1953; Hicks and Young, 1975). Recently Teja and his coworkers (1989) have developed two novel methods for the determination of the critical properties of thermally unstable fluids, i.e., a rapid heating sealed tube method and a low residence time flow method. The first can be used to measure the critical temperature and density while the second can be used to determine the critical temperature and pressure. These two methods provide the critical data with high accuracy for thermally unstable fluids due to the short residence time of the sample in the apparatus before the critical point is reached. For the critical temperature measurement of the thermally unstable fluids, the results from the flow method seem to be more accurate since the fluids are maintained at their critical temperatures for a short time period (ca. 10-50 s) which minimizes any thermal decomposition (Rosenthal and Teja, 1989). It is difficult to find suitable estimation methods for complex multicomponent mixtures such as jet fuels although a variety of correlation methods for estimating critical properties of petroleum fractions and coal fluids have been reported (Roess, 1936; Cavett, 1962; Kesler and Lee, 1976; Riazi and Daubert, 1980; Brule et al., 1982; Twu, 1984; API Technical Data Book, 1987). In this work the critical temperatures of nine jet fuels were measured using a sealed tube method (Hicks and Young, 1975; Mogollon et al., 1982) and the critical
temperatures and pressures of the fuels were estimated using the methods recommended in API Technical Data Book (1987). Five other methods, developed by Cavett (1962),Kesler and Lee (1976),Riazi and Daubert (19801, Brule et al. (19821, and Twu (19841, were also used to estimate the critical temperatures of the fuels. The measured and estimated critical temperatures were compared and the suitability of the estimation methods was discussed. In view of the shortcomings of these methods in estimating the critical temperatures of the coal-derived fuels, a new correlation was developed in the form of the Roess equation (API Technical Data Book, 1987)using the properties of the compounds found in the coal-derived jet fuels.
Experimental Section Samples. Six petroleum-derived jet fuels (designated as JP-8P, JP-8P2, Jet A, Jet A-1, JP-7, and JPTS) and three coal-derived jet fuels (JP-8C, JP-8CA, and JP8CB) were used in this work. The compositions of these fuels, except for JP-8CA and JP-8CB, have been reported by Lai et al. (1992). The petroleum-derived jet fuels are composed mainly of long-chain paraffins while JP-8C consists mainly of monocyclic and bicyclic alkanes and some two-ring hydroaromatic compounds. The compositions of JPSCA and JP-8CB are similar to that of JP-8C, although JP-8CA contains higher concentration of low-boiling-point compounds while JP-8CB is richer in high-boiling-point compounds. Critical Temperature Measurement. A sealed tube method was used to measure the critical temperatures of the jet fuels (Hicks and Young, 1975; Mogollon et al., 1982). The apparatus used in this work was constructed by Lyons (1985). The details of this equipment can be found in Lyons’ M.S. thesis (1985). In short, this method consists of sealing a certain amount of sample in a glass tube with a volume which is approximately equal to the critical volume of the sample at the critical temperature and then heating it rapidly in a furnace until the critical point is reached. The critical temperature is taken as the point at which the meniscus of the vapor and liquid disappears on heating or reappears on cooling through the critical point. If the loaded amount of the sample is such that the overall density in the tube is equal to the critical density, then the meniscus will disappear or reappear at the midpoint of the tube. On the other hand, if the overall density is
0888-5885/95/2634-0404$09.00/0 0 1995 American Chemical Society
Ind. Eng. Chem. Res., Vol. 34, No. 1, 1995 405 Table 1. Temperature Distributions along Axial Direction of the Furnace distance from thermocouple probe, cm
0 1
2 3 4
test 1 776 775 773 771 770
temperature, "F test 2 test 3 753 752 750 748 747
725 724 722 721 720
slightly larger than the critical density, then the disappearance or reappearance of the meniscus will occur slightly above the midpoint of the tube, or vice versa. There is no need for the overall density to be exactly equal to the critical density for an accurate measurement of the critical temperatures of the jet fuels since the precision within 1-2 O F is satisfactory for such complex mixtures consisting of hundreds of compounds (Lai et al., 1992). During the measurements it was found that there were some temperature variations along the axial direction of the furnace. For the sample tube used in this work, the position of thermocouple probe was about 3.5 cm away from the position of vapor-liquid meniscus observed just before the critical point was reached. Tests showed that this resulted in about 5 O F temperature difference between these two positions with higher temperature a t the position of thermocouple probe. Table 1 shows temperature distributions along axial direction of the furnace. Three tests gave similar temperature distributions, although different temperature ranges were used. Consequently the observed critical temperature was corrected by subtracting 5 O F from the thermometer readout. Simulated Distillation by Gas Chromatography. To estimate the critical properties of the jet fuels using the methods described in the following section, a volumetric average boiling point (VABP)must be known. The VABP can be calculated from ASTM D86 distillation data. In this work ASTM D86 distillation data were obtained from simulated distillation data by ASTM D2887, using the method described in API Technical Data Book (1987). In the simulated distillation experiment, a calibration mixture of hydrocarbons of known boiling points covering the boiling point range of the sample is run in a gas chromatograph and a calibration curve is obtained by plotting boiling point vs retention time. If the sample contains significant amounts of n-paraffins which can be identified on the chromatogram, these peaks can be used as internal boiling point calibrations. Since the jet fuel JP-8Pcontains identifiable n-paraffins from CS to c17, the peaks of these n-paraffins were used as internal boiling point calibrations. A mixture of npentane, n-hexane, and n-heptane was run t o cover the initial boiling points of the jet fuels. Figure 1 shows the relationships between the retention times of nparaffins and the corresponding atmospheric boiling points. M e r obtaining the calibration curve, the jet fuel samples were run in the gas chromatograph under the same operating conditions and the retention times a t 0,10,30,50,70,90, and 100 wt % points were obtained. Table 2 gives the retention times of the jet fuels at different weight percent points. The simulated distillation temperatures at the corresponding weight percent points were then determined using the calibration curve shown in Figure 1by converting the retention times to the corresponding boiling points. Table 3 shows the
0
5
10
15 20 25 30 Retention Time,min
35
40
Figure 1. Relationships between the retention times of nparaffins and the corresponding boiling points. Table 2. Retention Times (min) of Jet Fuels at 0, 10,30, SO, 70,90, and 100 wt % Points wt % distilled 30 50 70 90 100 0 10 fuel 11.3 15.7 20.2 27.1 38.7 JP-8P 3.3 7.0 JP-8P2 3.3 9.4 15.4 19.2 23.3 28.9 38.5 13.9 18.3 23.4 29.8 39.4 3.2 9.1 Jet A JetA-1 7.3 11.3 14.2 16.3 19.6 23.8 33.5 25.5 33.2 10.9 15.4 17.9 19.8 21.8 JP-7 JPTS 3.2 8.2 11.8 14.8 18.3 23.7 30.4 3.2 7.4 15.9 22.3 30.4 42.0 JP-8C 2.2 JP-BCA 2.1 3.2 7.0 13.7 19.6 28.7 39.7 9.6 16.4 23.8 31.2 41.3 JPdCB 2.2 3.8 Table 3. Simulated Distillation Temperatures (OF) of Jet Fuels at Different Weight Percent Points Obtained by ASTM D2887 wt % distilled fuel 0 10 30 50 70 90 100 489 595 385 424 JP-8P 242 304 345 242 326 382 416 453 506 593 JP-8P2 453 513 602 367 407 Jet A 240 324 390 418 457 548 345 370 Jet A-1 307 438 474 546 341 382 404 422 JP-7 350 376 407 456 520 240 316 JFTS 520 625 308 386 443 JP-8C 195 240 190 240 304 366 418 504 605 JP-8CA 527 620 390 457 JP-8CB 195 257 328
simulated distillation temperatures of the jet fuels at different weight percent points.
Critical Property Calculations The critical temperatures (T,)of the jet fuels were estimated by the method described in the API Technical Data Book (1987) and five other correlation methods, developed by Cavett (1962),Kesler and Lee (19761, Riazi and Daubert (19801,Brule et al. (1982),and Twu (1984). The critical pressures (P,)of the fuels were estimated only by the method described in the API Technical Data Book (1987) although other methods, except for the Brule method (Brule et al., 19821, can also be used to estimate this property. All the methods selected require only specific gravity and boiling point data as input. API Methods (API Technical Data Book, 1987). The API Technical Data Book (1987) recommends the Roess equation for Tcestimation:
T,= 186.16 + 1.6667A - 0.7127 A = SG(VAl3P
x 10-3A2 (1)
+ 100.0)
(2)
where T,is in degrees Fahrenheit, SG is specific gravity,
406 Ind. Eng. Chem. Res., Vol. 34, No. 1, 1995 Table 4. ASTM DS6 Temperatures (OF) of Jet Fuels at Different Volume Percent Points Converted from Simulated Distillation Temperatures at Corresponding Weight Percent Points vol % distilled fuel 0 10 30 50 70 90 100 JP-8P 332 352 379 298 409 463 539 JP-8P2 388 410 438 303 355 480 538 Jet A 300 352 374 401 438 486 546 Jet A-1 386 403 367 373 433 499 351 419 424 450 JP-7 384 403 407 499 JPTS 371 392 295 341 354 431 474 JP-8C 259 278 322 377 427 492 564 358 402 JPdCA 275 315 477 546 252 382 441 JPdCB 293 339 499 560 260
60 OF160 O F , and VABP is the volumetric average boiling point in degrees Fahrenheit which is the average of the ASTM D86 distillation temperatures at 10, 30, 50, 70, and 90 vol % distilled points. The ASTM D86 distillation temperatures were obtained from ASTM D2887, simulated distillation (SD) by gas chromatography, using the method described in API Technical Data Book (1987). Table 4 shows the ASTM D86 temperatures of the jet fuels at different volume percent points converted from the simulated distillation temperatures at the corresponding weight percent points. The P, was determined from Figure 4D 2.1 given in API Technical Data Book (1987) using the VABP, API gravity, and ASTM slope as parameters. The following five correlation methods were also used t o estimate the critical temperatures of the jet fuels. Cavett Correlation (1962).
T, = 768.07121
+ 1.713369316 - 0.10834003 x
10-2T; - 0.89212579 x lo-2(API)Tb -k
+
0.38890584 x 10-6T: 0.53094920 x ~O-~(API)T
