Atkins 2005
Atkins 2005
Atkins 2005
1935
Linda Shafer
University of Dayton Research Institute, 300 College Park, Dayton, Ohio 45469-0116
Received December 3, 2004. Revised Manuscript Received May 18, 2005
Introduction
Operators and designers of aircraft, both manned and
unmanned, that fly at altitudes above 12 000 m for
extended periods require high-fidelity fuel freezing
prediction tools. An impediment to the understanding
and simulation of jet fuel freezing has been the absence
of published thermophysical properties of jet fuel at low
temperatures. Common jet fuels, such as Jet A, Jet A-1,
and JP-8, have little published property measurements
at temperatures near their freeze point temperatures.
The freeze point for a hydrocarbon liquid fuel (ASTM
D2386-03) represents the temperature at which visible
wax crystals melt upon warming. Therefore, jet fuel
consists mostly of liquid several degrees (3-4 C)
below its freeze point. The absence of published property
data is likely due to the variation of jet fuels, based on
the composition of the original crude oil and the refinery
methods used. Also, property measurements at temperatures approaching the onset of solidification are difficult to obtain, requiring specialized equipment and
techniques. Commercial and military jet fuel specifications are not sufficiently rigorous to ensure consistent
thermophysical properties. Moreover, the introduction
of non-petroleum-derived fuels, such as Fischer-Tropsch fuels, are on the horizon. Therefore, jet fuel properties must be assessed regularly and new techniques
* Author to whom correspondence should be addressed. Telephone:
(937) 252-8878, ext. 114. Fax: (937) 252-9917. E-mail address:
jamie.ervin@notes.udayton.edu.
1936
Atkins et al.
Table 1. Volume Percentages of Species Classes within
Four Fuel Samples, as Determined by ASTM 2425
Composition of Jet A Fuel (vol %)
species classes
F3686
F3688
F3773
F3804
paraffins
cycloparaffins
dicycloparaffins
tricycloparaffins
alkylbenzenes
indan and tetralins
indenes, CnH2n-10
naphthalene
naphthalenes
acenaphthenes
acenaphthylenes
tricyclic aromatics
43.6
20.0
11.7
0.9
12.7
6.6
0.3
0.4
3.5
0.3
<0.2
<0.2
60.5
15.4
4.3
<0.2
13.7
4.3
<0.2
<0.2
1.5
<0.2
<0.2
<0.2
57.2
17.4
6.1
0.6
13.5
3.4
<0.2
<0.2
1.7
<0.2
<0.2
<0.2
58.5
15.7
3.7
<0.2
16.9
3.2
<0.2
<0.2
2.0
<0.2
<0.2
<0.2
100.0
100.0
100.0
100.0
total
F3686
F3688
F3773
F3804
aromatics
monoaromatics
diaromatics
total
total saturates
19.9
4.0
23.9
76.1
17.8
2.0
19.8
80.2
16.6
2.1
18.7
81.3
19.9
2.3
22.2
77.8
fuel
freeze point
((0.5 C)
cloud point
((1.0 C)
pour point
((2.0 C)
JP-8 (F3804)
JP-8 (F3773)
Jet A (F3686)
Jet A (F3688)
-48.3
-49.4
-42.0
-41.6
-52.6
-53.7
-46.2
-46.5
-58.0
-58.0
Density, F
((0.001 g/mL)
@ -20 C
@ -40 C
Kinematic Viscosity,
((0.5 cSt)
@ -20 C
@ -40 C
Viscosity,
((0.5 cP)
@ -20 C
@ -40 C
0.822
0.834
3.9
8.4
3.2
7.0
0.824
0.833
4.9
11.9
4.0
9.3
diaromatics, total aromatics, and total saturates. In lowtemperature studies of jet fuel, the freeze point temperature
is a parameter commonly used to characterize the fluidity of
the fuel. Table 3 lists both the measured freeze and cloud point
temperatures for the fuel samples (using a Phase Technology
Series 70V petroleum analyzer; ASTM D5972-96). For the
temperature regime in which fuel crystallization occurs, the
incipient solid crystals consist of nonideal solid solutions of
the larger n-alkanes (such as C16-C19) within the fuel.13 These
large n-alkanes represent a small fraction of the total n-alkane
concentration. Thus, the precipitation of n-alkanes from the
fuel (cloud point), and, therefore, the freeze point, are dependent on the n-alkane distribution of these species. Figure 1
shows the n-alkane distribution for the fuel samples considered. In addition, Figure 1 shows that fuels F3688 and F3686
have n-alkane distributions that have a greater fraction of the
larger n-alkanes (C16-C19). Therefore, it may be anticipated
that they would have the highest freeze point temperatures,
as demonstrated by Table 3.
The n-alkane analyses of the fuels were performed using
gas chromatography (Agilent model 6890) combined with mass
spectrometry (GC/MS) and flame ionization detection (GC/
FID). The same column model was used in both detectors
(0.25 mm inside diameter (ID) 30 m DB-5MS with 0.25-m
film). The GC temperature program involved an initial temperature of 40 C (2-min hold) followed by ramping (5 C/min)
to 200 C and then an increase (30 C/min) to 280 C. A
constant column flow rate of 1 mL/min and a 40:1 split ratio
were used. The injector temperature was 250 C. The mass
spectrometer, transfer line, and FID detector were held at a
(13) Zabarnick, S.; Widmor, N. Energy Fuels 2001, 15, 1447-1453.
1938
Atkins et al.
Table 4. Properties of Calibration Fluid LNF-45, as
Provided by the Manufacturera
temp
(C)
viscosity,
(cP)
kinematic viscosity,
(cSt)
density
(g/mL)
-30
-40
-45
31.79 ( 0.12
66.04 ( 0.10
101.8 ( 0.8
36.14 ( 0.18
74.59 ( 0.06
114.5 ( 1.0
0.8791 ( 0.0005
0.8857 ( 0.0009
0.8904 ( 0.0004
(1)
For a rotational viscometer, eq 1 and the equipment dimensions can be used to derive a relationship between torque (M)
and the rotor angular velocity (). Equation 2, which is called
the Margules equation,14 shows the relationship between the
viscosity, the dimensions of the viscometer, and the torque in
a rotational viscometer. Equation 2 also shows that the torque
is directly proportional to the rotor angular velocity for a
Newtonian fluid:
M)
4h
1
1
R2R R2S
(2)
60
24
12
-20 to -25 C
4.0
9.6
10.0
3.9
18.0
2.2
-30 to -35 C
3.0
13.3
7.0
5.3
14.0
2.8
-40 to -45 C
2.0
20.0
4.0
7.7
12.0
4.1
uncertainty
maximum
((%)
relative torque
-25 to -30 C
3.0
11.2
10.0
4.5
18.0
2.5
-35 to -40 C
2.0
16.2
5.0
6.3
14.0
3.4
-45 to -50 C
2.0
25.0
3.0
9.7
8.0
5.1
(3)
where slope (m) and intercept (b) of the linear fit are
determined. Relative torque measurements of the calibration
fluid are then correlated to (T), allowing any measured
relative torque of a sample fluid to be correlated with an
absolute viscosity.
An uncertainty analysis was performed to determine the
precision of the rotational viscometer instrument. Relative
torque was continuously measured from -20 C to -45 C for
rotor speeds of 12, 24, and 60 rpm. The experiment was
repeated four times at each rotor speed for the JP-8 F3804
sample. Table 5 summarizes the results of the uncertainty
analysis and shows that the instrument is most precise at
relative torque measurements >6%. Table 6 summarizes the
propagated error in the relative torque ratio calculation, which
is used in shear rate studies to demonstrate the direct
proportionality requirement of eq 2.
(15) Wilcox, D. C. Basic Fluid Mechanics; DCW Industries: La
Canada, CA, 1997.
(16) MacCoull, N. Lubrication 1921, 7, 65.
(17) MacCoull, N. Lubrication 1950, 36, 63.
(18) Walther, C. Erdoel Teer 1928, 4, 510.
(19) Walther, C. Erdoel Teer 1931, 7, 382.
(20) Wright, W. A. J. Mater. 1969, 4, 19-27.
torque ratio )
60/24
torque ratio )
60/12
torque ratio )
24/12
-20 to -25
-25 to -30
-30 to -35
-35 to -40
-40 to -45
-45 to -50
(10.8
(10.4
(7.6
(5.4
(4.5
(3.6
(18.4
(18.2
(14.3
(14.1
(12.2
(8.2
(20.6
(20.6
(15.7
(14.9
(12.6
(8.5
1940
Atkins et al.
1942
Atkins et al.
Figure 8. Relative torque measurements at rotor speeds of 12, 24, and 60 rpm for JP-8 (F3804) with 5000 mg/L of A4647 cooled
at a rate of -5 C/h.
Figure 10. JP-8 (F3804) in an optical cell with surfaces cooled to (a) the cloud point temperature (-53.0 C). With further
cooling, the surfaces reach -58.0 C and are held at that temperature for (b) 37.5 min, (c) 42.5 min, and (d) 47.5 min.
1944
Atkins et al.
Figure 12. Clouding behavior of JP-8 (F3804) with A4647 additive at surface temperatures of (a) -45 C, (b) -50 C, (c) -55 C,
and (d) -60 C.
Figure 14. Torque ratios versus temperature for varying shear rates at a cooling rate of -5 C/h. JP-8 (F3804) with 5000 mg/L
of A4647 additive was used.
Figure 15. Absolute viscosity of JP-8, Jet A, and JP-8 with additive samples obtained using the rotational viscometer and capillary
tube methods. The two symbols represent two capillary tube measurements, and the curves represent the viscometer measurements.
1946
Atkins et al.
(uy)
eff ) K
n-1
(4)
) liq 1 -
-B
A
(5)
) liq exp(Dwax) +
Ewax
xdu/dy
F4wax
xdu/dy
(6)
Acknowledgment. This material is based on research sponsored by the Air Force Research Laboratory,
under Agreement No. F33615-03-2-2347. The U.S.
Government is authorized to reproduce and distribute
reprints for Governmental purposes notwithstanding
any copyright notation thereon. The views and conclusions contained herein are those of the authors and
should not be interpreted as necessarily representing
the official policies or endorsements, either expressed
or implied, of the Air Force Research Laboratory or the
United States Government.
EF049683K
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: