NASA Technical Memorandum 79015

HIGH-FREEZING-POINT FUELS USED

FOR AVIATION TURBINE ENGINES

OO

ss:*=c.oo

Robert FriedmanLewis Research CenterCleveland, Ohio

30 J.4NMCDONNELL DOUGLAS

RESEARCH & ENGINEERING LIBRARYST. LOUiS

TECHNICAL PAPER to be presented at theTwenty-fourth Annual International Gas Turbine Conferencesponsored by the American Society of Mechanical EngineersSan Diego, California, March.11-15, 1979

https://ntrs.nasa.gov/search.jsp?R=19790007028 2018-05-22T22:03:16+00:00Z

HIGH-FREEZING-POINT FUELS USED FOR AVIATION TURBINE ENGINES

Robert FriedmanNational Aeronautics and Space Administration

Lewis Research CenterCleveland, Ohio 44135

00OS

ABSTRACT

Broadened-specification aviation fuels could beproduced from a greater fraction of crude sourcematerial with improvements in fuel supply and price.These fuels, particularly those with increased finalboiling temperatures, would have higher freezing tem-peratures than current aviation turbine fuels. Thehigher-freezing-point fuels can be substituted in themajority of present commercial flights, since tempera-ture data indicate that in-flight fuel temperaturesare relatively mild. For the small but significantfraction of commercial flights where low fuel tempera-tures make higher freezing-point fuel use unaccept-able, adaptations to the fuel or fuel system may bemade to accommodate this fuel. Several techniquesare discussed. Fuel heating is the most promisingconcept. One simple system design uses existing heatrejection from the fuel-lubricating oil cooler, anoth-er uses an engine-driven generator for electricalheating. Both systems offer advantages that outweighthe obvious penalties.

INTRODUCTION

It is likely that aviation turbine fuels willcontinue to be conventional hydrocarbon types for anumber of years (1). The world-wide reserves ofpetroleum appear limited, demands for competing oilproducts are increasing, and of course prices haverisen rapidly. Fuel costs are now an appreciablefracture of commercial airline operating costs. Thesupply, competition, and perhaps the cost situationcan be eased if aviation turbine fuel specifications

Numbers in parentheses designate references atend of paper.

are broadened to include petroleum fractions with ahigher final boiling point and less compositional re-strictions, giving producers wider flexibility infuel refining (2). Relaxed specification jet fuelsmay also permit effective use of lower-grade crudesources, including shale oil, tar sands, and coalliquids with minimal refinery upgrading (3).

The optimization of future fuel characteristicsrequires an understanding of the relationship betweenairframe and engine requirements and fuel specifica-tions. In this respect, fuel freezing point is a keyproperty, affecting the fuel flow and pumpability athigh-altitude, long-range cruise conditions. Incipi-ent fuel freezing problems could certainly affectapparent engine performance. Fuel freezing-pointspecifications are sensitive to the final boilingpoint and blending composition of the fuel. In fact,nearly half the respondents in a recent refinery sur-vey (4) stated that freezing point was the key, orlimiting, property in their jet fuel production.Relaxed specification fuels with higher final boilingtemperatures can be expected to show higher freezingpoints. For example, the same survey data indicatedthat increasing final boiling temperature by 13° Cpermits a 20 percent increase in jet fuel production,but at the expense of an 8° increase in freezingpoint.

It is recognized that broadened-specificationfuels will introduce several important propertychanges affecting engine performance, efficiency, ordurability. However, this paper is concerned onlywith freezing point and the practical application ofhigher-freezing-point aviation turbine fuels. Themeasurement and significance of freezing point is dis-cussed first, since the low-temperature behavior ofhydrocarbon fuels is not always uniquely described bya single laboratory characteristic. The temperature

environment of present commercial airplane missionsis then defined from statistical and calculated data.Finally, methods of adapting higher-freezing-pointfuels to the needs of current commercial airplanesare reviewed.

MEASUREMENT OF JET FUEL FREEZING POINTS

Specification TestsBecause aviation turbine fuels are complex mix-

tures of many compounds, freezing or crystallizationdoes not occur at a definite temperature, but a phasechange is exhibited over a range of temperatures.The higher molecular weight, straight-chain or normalparaffins (and certain symmetrical molecule hydro-carbons) crystallize first in the form of waxes. Astemperatures are lowered a waxy matrix builds up,trapping other liquid constituents of the fuel untila nearly solid structure is formed (5). Severalfreezing point specification tests have been devised,based on the temperature at which wax crystals firstappear, which in effect defines a conservative, maxi-mum freezing temperature. These tests include waxappearance, cloud point, and wax disappearance methods(table I). Wax appearance is the observation of thetemperature of the first appearance of the crystalsin a chilled, stirred fuel sample. Cloud point isthe temperature at which a solid suspension or "cloud"is observed near the bottom of an unstirred sample.Wax disappearance temperature is the temperature ob-served at the disappearance of crystals in a stirredfuel sample remove'd from a freezing bath and warmedafter the first crystals are formed. In contrast, aminimum, solid phase, "freezing" temperature may bedefined by the pour point, which is the lowest tem-perature at which the surface of a chilled fuel sam-ple will move when turned vertically. Table I alsolists approximate values of precision and accuracyfor these test methods. For each specification,table I lists estimates for: (1) repeatability, themean deviation between duplicate measurements by thesame observer; (2) reproducibility, the mean devia-tion between comparison measurements by different ob-servers, and (3) bias, the mean deviation of thespecification with respect to wax disappearance spec-ification D-2386-67, which is adopted as the presentstandard for aviation turbine fuels because of im-proved precision. The specifications (except forD-1477-65, which is now obsolete) are described indetail in the latest edition of the American Societyfor Testing and Materials (ASTM) Standards (6).Repeatability, reproducibility, and bias estimatesare those published in this reference or reported ata December 1976 low-temperature symposium of theASTM.

Freezing point measurements as a function ofboiling point are illustrated in figure 1. The cir-cular data points are freezing points^of narrow boil-ing range (8° to 14°)2 fractions of a*Diesel-typefuel, measured at the NASA Lewis Research Center.The difference between wax appearance and wax dis-appearance temperatures are represented by the shadedband, which ranges from 0° to 4°, being greatest atthe lowest temperatures where the absolute accuracyof the measurements is lowest. (For acceptance test-ing, specification D-2386-67 requires that the differ-ence between wax appearance and disappearance be nogreater than 3° or the test must be repeated.) Datafor wax disappearance tests of two other fuels arealso shown in figure 1: a conventional petroleum-

Temperatures are reported in degrees Celsius(°C) throughout this paper.

derived aviation turbine fuel, and a similar fuel de-rived from shale oil (7). The three fuels shown infigure 1 represent a wide variation of composition.Freezing points vary as much as 15° at the same boil-ing temperature. In general, above a boiling tempera-ture of 230°, the data show that freezing point in-creases with boiling temperature as expected, with anapproximate slope of 0.6 degree per degree. Below230°, the freezing point-boiling temperature relation-ship is not consistent; and over some temperature in-tervals, freezing point is seen to decrease with in-creasing boiling temperature.

Aviation turbine fuels are of course petroleumfractions with a range of boiling temperatures ratherthan a single boiling point. The higher boiling mate-rial is richest in the higher-molecular weight frac-tions, which tend to have higher freezing points,especially for the normal paraffins. Hence the maxi-mum (final) boiling temperature is generally far moreinfluential in characterizing the freezing point thanthe minimum (initial) boiling temperature. In addi-tion, it is likely that broadened-specification avia-tion turbine fuels will have increased final boilingtemperatures but unchanged initial boiling tempera-tures, with respect to present specifications, toavoid problems of decreased flash point and increasedvolatility (8). Future trends of freezing point maythen'be represented, as discussed in the INTRODUCTION,as a function of a single boiling temperature, thefinal boiling point (1,4). Figure 1 may also be in-terpreted as an illustration of the freezing point-final boiling temperature relationship.

Performance TestsPerformance-oriented freezing-point measurements

relate the low-temperature behavior of a fuel to adesired operating characteristic. These techniquesmay include measurement of viscosity-temperature rela-tions, filter or pumping pressure drop, and flow ratethrough tubes or capillaries under pressure or vacuum.Most of the methods were devised for testing Dieselor fuel oils for winter service (9,10), but they mayhave eventual application to jet fuels.

At least two methods have particular interest forstudies of low-temperature performance of aviationturbine fuels, A liquid-solid apparatus of Dimitroff(5) used a 100 kPa (one atm) pressure differential toforce liquid through a filter in a cold, constant-temperature bath for two hours. The difference infuel volume between the original sample and the liquidrecovered through the filter is the fraction crystal-lized. A cold flow test developed by Shell Research,Ltd., measured the gravity flow discharge of chilledfuel between two chambers separated by a valve (11,12).The solid fuel retained in the upper chamber is re- 1corded as a percent holdup. For both these perfor-mance tests, measurements can be taken at severaltemperatures to yield a plot of fraction of fuel fro-zen or held up as a function of temperature.

Large-scale flow tests simulating actual serviceconditions for aviation fuels were also conducted byShell Research (12) in an apparatus consisting ofrectangular or contoured aircraft fuel tanks mountedin a cold room. Fuel in the tanks was cooled for 32to 72 hours to a uniform.test temperature and thenpumped out of the tanks to an external receiver. Thefraction of fuel remaining in the tank that cannot bepumped was called the holdup, and several tests estab-lished a relation of holdup against temperature,analogous to the aforementioned smaller-scale testtechnique.

Example of results of performance tests, deter-mining fraction of frozen fuel as a function of tern-

temperature, are shown in figure 2. Tests of threefuels by the Shell cold tank method (data from (13)and additional data from Dr. Peter Ford) and of afourth fuel by the liquid-solid separator (5) areillustrated. Fuel A, a jet fuel with a high normalparaffin content, has a narrow range of freezing be-havior, about 1.5° from liquid to completely solid.In contrast, fuel C with a low n-paraffin content,shows a wide range of freezing behavior; in fact, itis limited to only 10 percent solid by the -60° lowtemperature capability of the apparatus. Fuel D is awide boiling range oil, covering jet and Diesel spec-ifications, also with a low n-paraffin content. Thefreezing behavior of this fuel was measured by theliquid-solid separator, yet the slope of the freezingcurve for this fuel in figure 2 agrees well with thatof fuel C measured by holdup tests. Finally, curve Billustrates results for an intermediate n-paraffincontent fuel, with a range of freezing behavior cover-ing about 8°. Laboratory studies of freezing behaviorof mixtures of known composition confirm that normalparaffins crystallize most readily, and accordingly,the high n-paraffin fuel will form a complete crystal-line matrix over a short temperature span. The lown-paraffin fuels, in contrast, exhibit a much slowerbuildup of the solid matrix extending over a largerrange of temperatures. The zero holdup temperaturesmeasured by the small scale cold flow tests (13) agreewith the corresponding temperature measured in thefull scale cold tank tests. On the other hand, thespecification wax disappearance freezing point doesnot necessarily correlate with the zero cold tankholdup, nor does pour point correlate with the 100percent holdup. These discrepancies are especiallyevident for higher boiling fuels above the jet fuelrange.

FUEL TEMPERATURES IN COMMERCIAL AIRPLANES

In-Flight Fuel TemperaturesThe fuel temperature in aircraft tanks varies

with the rate of heat transfer to the surrounding en-vironment. The heat flux, largely convective, is afunction of the total air temperature, or more correct-ly the turbulent boundary layer recovery temperature.

TR = (1 + 0.18M2) Ts

where TR is recovery temperature, Tg is static tem-perature, and M is Mach number. For a present daycommercial aircraft cruising at a Mach number of 0.84,the recovery temperature at a static air temperatureof -72° is -46°. The difference between recovery andtotal temperature is only 2° to 3° at these conditionsand, therefore, no distinction will be made in thispaper between recovery and total air temperatures.

In-flight measurements showing fuel and total airtemperature for two commercial wide-bodied aircraftare presented in figure 3. The solid lines are fueltemperatures measured by a single thermocouple in thefuel tank. The broken lines are the correspondingtotal air temperatures measured by external thermo-couples. The data for the 4400 km commercial flight,furnished by the Lockheed Company, indicate that thefuel cooled porportionately to the total air tempera-ture, but fuel temperatures remained approximately10° above the total air temperature throughout theflight. The very long range flight data were recordedby Pasion and Thomas (14) based on a Boeing Companydelivery flight. The fuel temperatures closely ap-proached the total air temperature during portions ofthe flight. When the airplane traversed zones of

rapidly changing total air temperature, however, thefuel temperatures responded much more slowly. Mini-mum altitude static temperature, not shown in figure3, were reported as -67°.

Calculated Fuel TemperaturesThe predictive calculation of fuel temperatures

supplements the in-flight measurements and establishesparametric variations and extreme statistical proba-bilities. Fuel temperatures are calculated from asummation of time-related heat fluxes over an assumedschedule of total air temperatures for a flight peri-od. Several computing programs of this type havebeen developed. A Boeing Company program (15), usedfor the calculations presented in this paper, includesthe effects of external convective and radiative heatfluxes and internal heat transfer from the pump, hy-draulics, and intertank fuel transfer. The majorsimplification of this calculation is the assumptionof a uniform fuel temperature within the fuel tank.Overall gradients of 10° from top to bottom have beenmeasured in aircraft fuel tanks and in simulatedtesting apparatus during cooldown. However, thesegradients do not have a great effect on the accuracyof the computed bulk fuel temperature. A maximumdiscrepancy of 2.5° was reported for comparisons ofcomputed temperatures with a variety of correspondingfuel temperature measurements (15).

Two examples are presented of calculated fueltemperature histories at extreme ambient conditionswith a one-day-a-year (0.37.) probability (14). Fig-ure 4 shows results for short range, 900 km flight,based on a -60.7° static air temperature and a flightschedule of climb, cruise at a Mach number of 0.82,and descent. Two cases are shown: one for the fuelloaded at 12°, the other for fuel loaded at -21°.For the colder initial fuel case, the fuel chilledcontinuously during flight, reaching -31.5° at theend of the 92 minute flight. For the warmer case,the fuel cooled more rapidly, but the fuel tempera-ture was still about 10° above that of the cold ini-tial fuel case at the end of the flight. These his-tories resemble the measured data for the 4.7 hourLockheed flight shown in figure 3. It is thus shownthat, for short range flights, there is not enoughtime for the fuel temperature to cool to the totalair temperature, and fuel temperatures throughout theflight retain the influence of the initial loadingtemperature.

Figure 5 shows results for a long range flight,9100 km. The total air temperature schedule is basedon a flight profile with cruise at a Mach number of0.84 and an altitude static temperature variationthat includes a portion of the flight at a minimumtemperature of -72°. Calculations for initial fueltemperatures of 12° and -21° are shown. After aboutsix hours of flight time, there is no longer an in-fluence of initial temperature. The fuel temperatureeventually approaches the total air temperature, anda minimum fuel temperature of -43° is predicted.This type of fuel temperature history resembles themeasured data for the long range Boeing flight shownin figure 3. For long range flights, fuel is exposedto low temperatures long enough that the fuel tempera-tures can approach the total air temperature, and theminimum fuel temperature is independent of the ini-tial loading temperature. It should be noted thatthe minimum fuel temperature does not necessarilyoccur at the same time of flight as the minimum totalair temperature because of the slow response of thefuel temperature to external temperature changes,during the coldest portions of the flight.

Freezing Point Requirements for Commercial FlightsIn-flight problems with freezing fuel have not

been documented. Situations have occurred, however,where gage warning of low fuel temperatures haveprompted flight crews to divert to warmer, low alti-tude air masses. Such diversions are costly in termsof operation at less than optimum altitude-speed com-binations or increased path length. A tolerance of3° to 5° above the specification freezing point isrequired as a minimum fuel temperature during flight.It has been pointed out that the sum of the reproduc-ibility of the freezing point specification, the ac-curacy of the thermocouple and gaging system, andother factors can add up to uncertainties much greaterthan these tolerances, but experience has proved thateven 3° is an adequate margin of safety.

It is interesting to note that for the calculatedextreme case flight shown in figure 5, a fuel wouldbe required with a maximum freezing point of -46°(-43° minimum temperature minus the 3° tolerance).Fuel meeting the aviation turbine fuel standard spec-ification D-1655-77, type Jet A-l, has a maximumfreezing point of -50°. Jet A-l, however, is no long-er produced in the United States. The common domesticaviation turbine fuel, type Jet A, has a maximum spec-ification freezing point of -40° although Departmentof Energy inspection statistics (16) report that thearithmetic average of the freezing point of Jet Asamples in 1977 was -45°.

ADAPTATIONS FOR USE OF HIGHER-FREEZING-POINT FUEIS

Higher-Freezing-Point FuelsAs discussed earlier, fuels with increased freez-

ing point specifications could make a greater propor-tion of jet fuel available in overall production, andthese fuels may indeed be a necessity in the nearfuture. Although a calculated extreme case shown infigure 5 indicates that there can be flights where afuel with a freezing point of -46° or lower is neces-sary to meet specification tolerance limits, the major-ity of commercial flights have milder freezing pointrequirements. Pasion and Thomas (14) estimate that a'fuel with a -29° freezing point would be acceptablefor use in all flights up to 9100 km in length insummer, and in 94 percent of these flights in winter.The small fraction of remaining flights cannot beignored, of course. These long-range flights, mostlyinternational, could not use a higher-freezing-pointfuel except with costly, low-altitude operation. Thespecification of several grades of jet fuel is an ob-vious suggestion, but this would be objectionable toboth suppliers and users because of complications inproduction, distribution, and storage. The most prom-ising approach for the future would be to specify asingle grade of higher-freezing-point fuel, to be usedfor all commercial flights, with certain fuel or air-frame adaptations made to meet the needs of the long-range flights. The most likely or practical modifica-tions are discussed in this section.

Fuel Modifications

Preheating. For short-range flights, ground pre-heating of fuel can indeed keep the fuel warmerthroughout the flight (fig. 4). Unfortunately, forthe long-range flights where fuel temperatures aremost critical, initial fuel temperatures have littleor no influence in flight (fig. 5). Ground preheatingof fuel prior to long-range flights may be helpful forstorage and ground handling of very high-freezing-point fuels at extreme winter conditions (14).

Additives. It was previously pointed out thatthe freezing process of fuels generally involves abuildup of a waxy or crystalline matrix, which even-tually traps even liquid constituents as gels (5,17).Certain polymeric compounds in small concentrationsare effective in delaying the matrix buildup by dis-persing the initally formed solid particles. Theseadditives are pour point depressants, which improvethe low-temperature flow properties of the fuel with-out affecting the crystal formation as observed byfreezing or cloud point tests. The mechanism of thisdispersion probably involves several physical actionsaltering crystal size, shape, and surface attractiveforces (18). The flow-improving additives are oftenblended into lubricating or fuel-oils for winter ser-vice, but they are not effective in modifying the low-temperature properties of jet fuels. The additivesmay be of interest, however, in the production offuture broadened-specification jet fuels because thehigher boiling fractions blended into these fuels maybe quite sensitive to additives (19,20). In figure 6,an example of this behavior is shown. These data aretaken from Shell cold tank tests (13), and the percentof frozen fuel held in the tank is plotted as a func-tion of temperature. The base curve, shown as a solidline, is the freezing behavior of fuel B in figure 2.The freezing results of a wider-boiling-range fuelare shown by the broken curve in figure 6. This fuelwas a mixture of the base jet fuel plus 20 percent ofa heavier, higher-final-boiling point fraction fromthe same source material. The low temperature per-formance of the wider-boiling blend is poorer thanthat of the base fuel as evidenced by greater holdup,or less flowability, at the same temperature. Thesecond broken curve shows the performance of the sameblend with the addition of 300 ppm of a flow improver.In this case, the holdup of the wide-boiling fuel hasimproved to the extent that it surpasses that of thebase fuel.

Greater utilization of additives awaits specifica-tion standards based on low temperature flow charac-teristics, rather than wax disappearance freezingpoints insensitive to flow improvers.

Aircraft Fuel System Modifications

Insulation. Fuel heat losses to the surroundingenvironment are reduced by insulating the fuel tank.Pasion and Thomas (14) calculated the effect of arealistic system of insulation that could be addedexternally to the wing surface of a wide-bodied air-,plane. A 2.5 cm thickness of an epoxy filled glassmicroballoon layer was shown to reduce fuel heatlosses by about 20 percent. The calculated fuel tem-perature history for a long-range flight with insu-lated fuel tanks is shown in figure 7, compared tothe reference uninsulated case from figure 5. Theinsulated fuel is about 6° warmer than the uninsulatedfuel during most of the long-range flight, but thedifference in minimum temperature is about 4.5°. Theinsulated fuel does not reach the minimum temperatureuntil after 9 hours, at the end of the flight. Fueltank insulation, however, adds serious structuralproblems and severe weight penalties even for simplesystems. It is, moreover, a constant penalty in con-trast to heating systems which can be turned off whennot used.

Fuel heating systems. Existing airframe and en-gine heat rejection sources offer the most promisingmethod of maintaining fuel above minimum temperaturesin flight. Calculated fuel temperature histories forheated fuel (14) are shown in figure 8, which in-

eludes, for comparison, the reference calculated long-range temperatures (fig. 5) for unheated, uninsulatedfuel tanks. The middle solid curve shows the resultof computations which assume that 79 kW of heat perengine is added continuously during flight. Thesecalculations are based on a four-engined wide-bodiedairplane, in which each engine is fed by the equiva-lent of a single fuel tank. The heating rate of 79 kWis equivalent to the energy released by the combustionof 6.5 kg of fuel per hour, which is a negligiblefraction of a typical, efficient engine cruise fuelconsumption of about 2000 kg/hr. This constant fuelheating rate will maintain the minimum fuel tempera-ture above -25°. Continuous heating is not necessary.The heating system could be turned on when the fuelis 5° above the desired minimum temperature, for in-stance, and heating would still be effective, sincefuel cooldown rates are relatively slow during theportion of the mission where temperatures are lowest.If the fuel tank were insulated with the previouslyassumed 2.5 cm thick material, fuel temperature wouldrise slowly throughout the flight with this rate offuel heating, as indicated by the highest solid curvein figure 8. Combined fuel heating and insulationmay be used to reduce the heat requirements. Theminimum temperature of -25° can be maintained by 20 kWof heating if the fuel tank is insulated, as shown bythe broken curve in figure 8.

Pasion and Thomas (14) also evaluated methods anddesigns for heating fuel in flight. Figure 9 is asketch of a simple system using the existing heat re-jection of the engine by the addition of a secondlubricating oil heat exchanger in series with theexisting lubricating oil cooler. In the ordinarymode of operation, the fuel control relieves excessfuel to the engine fuel pump. For fuel heating, athree-way valve diverts the excess fuel through theadded heat exchanger into the fuel tank to warm thebulk fuel. The heating system does not affect theoperation of the fuel control or engine fuel system,whether in the normal or fuel heating modes of opera-tion.

The calculated fuel temperature for a long-rangeflight with the lubricating oil-fuel heating systemis shown in figure 10. The heating rate schedule,calculated by the engine manufacturer, includes theheat available from the lubricating oil and the smallamount of heat rejection from the pump and fuel con-trol. Although heating rates vary, particularly overthe first climb phase of the flight, they are reason-ably constant at about 50 kW throughout most of thecruise portion of the flight. The minimum fuel tem-perature is about -31°, which shows that this simplesystem will permit use of a broadened-specificationfuel with a -34° freezing point (including the 3°operating tolerance). Weight, performance, and modi-fication cost penalties of the added system are small.On the basis of preliminary calculations (14), evensmall fuel cost reductions through the use of broad-specification fuels can readily offset these penal-ties. The heating of the fuel tank by engine heatrejection is not only a promising concept for heatinghigh-freezing-point fuels, but it may also prove ad-vantageous for engine designs with problems of exces-sive heat rejection, such as those with reductiongearing (21).

A second approach to in-flight fuel heating isthrough the use of modifications that divert a smallportion of the engine energy to fuel heating, in con-trast to the use of existing heat rejection. A prom-ising example is an engine-driven electrical generatorto heat fuel electrically, using a secondary heattransfer loop for control and safety (14). There are

present day military aircraft with functional engine-driven auxiliary generators, although the installationsare for electronic demands rather than for fuel- heat-ing. The advantages of engine-driven electrical heat-ing are that it is simple and efficient; it can bediverted for ground heating or other uses; and it cangenerate practically any amount of constant heatingpower desired. The disadvantages of this system overthe use of existing heat rejection, the lubricatingoil cooler for example, are in the greater mechanicalcomplexity of the engine-driven system; the small butappreciable engine thrust penalty; and the greater in-stallation and operating costs.

CONCLUDING REMARKS

If aviation turbine fuels were specified withless compositional restrictions and higher final boil-ing temperatures, they could be produced from a great-er fraction of the crude oil sources with obvious ad-vantages of efficient utilization of raw material,reduced refinery complexity and energy consumption,and perhaps lower product cost. The broadened-specification fuels would most likely have severalproperty changes of serious concern to engine and air-frame technologists, but this paper concentrates onfreezing point, which generally increases as boilingtemperature increases.

This paper has reviewed the freezing behavior ofpresent and broadened-specification jet fuels. Per-formance and flow tests indicate that, in many cases,fuels have a usable flow or pumpability at temperatureslower than the standard freezing point defined by awax disappearance test. Higher boiling-point fuelsare also sensitive to flow-improving additives whichcan extend the range of low temperature flowability.As yet, no standard low temperature flow specificationhas been accepted that defines this usable low-temperature flow.

In recent years, airlines have compiled extensivedata on in-flight fuel temperatures. For most mis-sions, the fuel temperature environment is mild, andpresent freezing-point specifications of -40° C maxi-mum could be relaxed. For example, data show that a-29° C freezing point jet fuel may be used in allflights up to a range of 9100 km in summer and in 94percent of these flights in winter. Moreover, thisbroad specification fuel may be used for all flightsunder any weather conditions if certain adaptationscan be made. Heating the fuel tanks in flight appearsto be the most practical approach for general use ofhigher-freezing-point fuels. Two promising systemsuse engine lubricating oil heat for low power, smallpenalty fuel heating, or engine-driven electrical heatfor high-power heating at greater penalties. Theseconcepts are still in an early state of development.

REFERENCES

1 Longwell, J. P., and Grobman, J., "AlternativeAircraft Fuels," ASME Paper No. 78-GT-59, Apr. 1978.

2 Robertson, A. G., and William, R. E., "JetFuel Specifications: the Need for Change," ShellAviation News. No. 435, 1976, pp. 10-13.

3 Dukek, W. G., and Longwell, J. P., "Alterna-tive Hydrocarbon Fuels for Aviation," Exxon Air World,Vol. 29, No. 4, 1977, pp. 92-96.

4 Dickson, J. C., and Karvelas, L. P., "Impactof Fuel Properties on Jet Fuel Availability," AFAPL-TR-76-7,.Apr. 1976, Air Force Aeropropulsion Lab.,Wright-Patterson AFB, Ohio. (AD-A029493).

5 Dimitroff, E., et al., "Crystal-Liquid FuelPhase Intersolubility and Pumpability," Seventh WorldPetroleum Congress Proceedings, Vol. 8, Part 2,Elsevier, New York, 1967, pp. 141-155.

6 "Petroleum Products and Lubricants, Parts 23-(I), 24-(II), 25-(III)," Annual Book of ASTM Standards,American Society for Testing and Materials, Philadel-phia, 1978.

7 Solash, J., Nowack, C. J., and Delfosse, R. J.,Evaluation of a JP-5 Type Fuel Derived from Oil Shale,"NAPTC-PE-82, Mary 1976, Naval Air Propulsion Test Cen-ter, Trenton, New Jersey. (AD-A025417).

8 Longwell, John P., ed., "Jet Aircraft Hydro-carbon Fuels Technology," NASA CP-2033, Jan. 1978.

9 Dodson, S. C., and Fairman, A. G. C. "FlowProperties of ATK at Low Temperature," The Engineer(London), Vol. 21, No. 2488, Mar. 31, 1961, pp. 499-503.

10 Coley, T., Rutishauser, L. F., and Ashton,H. M., "New Laboratory Test for Predicting Low-Temperature Operability of Diesel Fuels," Journal ofthe Institute of Petroleum, Vol. 52, No. 510, June1966, pp. 173-189.

11 Strawson, H., "The Pumpability of AviationTurbine Fuels at Low Temperatures," Journal of theInstitute of Petroleum. Vol. 45, No. 425, May 1959,pp. 129-146.

12 "Proposed IP Standard Method for Cold FlowTest of Aviation Turbine Fuels," Journal of the In-stitute of Petroleum, Vol. 48, No. 467, Nov. 1962,pp. 388-390.

13 Ford, P. T.s and Robertson, A. G., "Jet Fuels -Redefining the Low Temperature Requirements," ShellAviation News, No. 441, July 1977, pp. 22-26.

14 Pasion, A. J., and Thomas, I., "PreliminaryAnalysis of Aircraft Fuel Systems for Use with Broad-ened Specification Jet Fuels," NASA CR-135198, June1977.

15 Barr, N. M., et al., "Boeing Airplane FuelSystems at Low Temperatures," Document D6-42386, 1975,Boeing Commercial Airplane Co., Seattle, Wash.

16 Shelton, E. M., "Aviation Turbine Fuels,"BERC/PPS-78/2, May 1978, U.S. Department of Energy.

17 Dimitroff, E., and Dietzmann, H. E., "FuelCrystallization-Additive Behavior and Compatibility,"American Chemistry Society, Division of PetroleumChemistry. Preprints. Vol. 14, No. 3, Sep. 1969,pp. B132-B143.

18 Button, J. F., "Flow Properties of Distillatesat Low Temperatures. A Review," Journal of the In-stitute of Petroleum. Vol. 45, No. 425, May 1959,pp. 123-129.

19 Knepper, J. I., and Hutton, R. P., "Blend forLower Pour Point," Hydrocarbon Processing, Vol. 54,No. 9, Sep. 1975, pp. 129-136.

20 Ball, K. F., "More Jet Fuel from Heating Oil,"Hydrocarbon Processing, Vol. 56, No. 9, Sep. 1977,pp. 122-124.

21 Coffinberry, G. A., and Kast, H. B., "Oil Cool-ing System for a Gas Turbine Engine," U.S. Patent4,020,632, May 3, 1977.

oCOCO CJ•rl OPP

4-1•rl

•rl43 CJ•rl OJJ M

COcucxcu

CO

4J•rlr-l•rl43 CJ•rl OJj II

COCUcx0)

cucx

£_l

cu1-14-1•rlH

,oa4JCO0)H

111

vO•

CN

•o

1cxCOCO•rlFTJ

XCO

&

o4J0•rlOCX

IN01o>(_l

Cu

vO1

vOooCOCN

1P

cuCJ0CO^4COCUCX

COi— icu3

14-1

0O

•i-l4JCO

Ir4

CO

CM1

CN•

CM

00

*

O

1r!CO0)CXCX 0)CO O

pjX coCO13

1-1•H4-1

0) COO -rl0 T3 coCO r-lr4 IK CUco O 3CU t*-lCX 4JCX 0 CUCO -rl 4J

O COX CXr-lCO13

CNp^

11 ,r-l1— 100

1Q

CM1

r-l•

CO

oo•

0

1

COcucxCX 01CO CJ

0X coCO!3

O COI-l

4J CUa 3o cucx 0 *->

o cu60-rl r-lC 4-1 O

•rl CO CON <rl 43cu > o0) co >~ 'l

Pn

mvO

1f^p^j-

r-l1p

CO1

^J-

CN

COcucxcxCO

XCOs

M-lo4J0

•1-1ocx

•n3or-lCJ

vOSO

1o0mCN

1p

0)CJ0CO

COI-l•rloEiCUi-4orl4Jcucx

001

-a

vO

CO

X•rlrl4JCOeXCO12

IWo4-1c•rlocxrl

3oPM

vOvO

1f 1**c^p

COr-4•Hoes

1— 1orl4-)0)cx

v

y-^vO^^

r4curl

cuCO

ocugCOCO

KS

43

CO4-10cucurl

3COCOcu

4JCOcucxcurl

0cucu1 1

cuft

cuo

I01

14-114_|•rlp

CO .

•

^^VO

COrl

01

^4CUCO

43o4J0curl0)

M-lM-l•rl""0

>!

CO4J0CU

curl

3COCO0160COCU

B01

43

cuogrl0<

14-114-1•rloa

43

T)cu4JrlocxCUrl

COCO

0)o0

. cui-l0)

curl

COCO

J^^vO

1

00COCN

pCO30

•H. eT)orl4J

taM-Jo0)CJ001rlO)

M-l

>P4

T3

cu4-1CO

•rl

orlcxa

o

re

Requirem

ents

34JCOrl

0)cxcuH15o

g_3

r-l013

4J0)

1— 5

U-logrj

•HCOoCXip

CO

5J§H

CO•

0)

54JCO

Jrf01m3p

*a•

ES

or-l

S-l0)4JCO01rl

60

rl

oor^1-1

1

•o0CO

o0

00)0>[54J0)

43

01

0°COJ_J

2gCQcu3r-l

CO

. s-\ COvO COr^ *rlO\ 43r-4

cu

J cu0) rlP 4J

w*o

o00oI

w

lOi—

-60

O DISTILLATE (DIESEL) FUEL - WAX DISAPPEARANCE• DISTILLATE (DIESEL) FUEL - WAX APPEARANCED JET A - WAX DISAPPEARANCEO SHALE DERIVED JP-5 (REF. 7) - WAX DISAPPEARANCE

180 200 220 240 260 280 • 300 320 340 360B O I L I N G POINT. °C

Figure 1. - Freezing points of narrow boiling range samples offuels.

MEASUREMENT BY TANK HOLDUP(REF. 13)

MEASUREMENT BY LIQUID-SOLIDSEPARATION (REF. 5)

FUEL NORMAL PARAFFINCONTENT,percent

42.5301521.5

-55 -50TEMPERATURE, °C

'1 I-45 -40

Figure 2. - Fuel freezing behavior.

o

§ o>

T s*^If03 t/1

CJ

« . = •^ JZ fe

£asis

DO

DO '3anivd3dwn

oCOoI

H

20 1—

10

0

-10

-20

-30

-40

-50

i^FUELLOADEDAT12°C

-FUEL LOADED AT-21° C

-ASSUMED TOTAL AIRTEMPERATURE VARIATION

0 4 5TIME, hr

Figure 5. - Predicted fuel temperature for a long-range (9100km) commercial aircraft mission, based on a minimumstatic air temperature of -72° C.

100

80

fe 60

o 40

20

-JET FUEL

-JET FUEL PLUS 20% HIGHER/' BOILING MATERIAL

FUEL, PLUS 20!!.

\ \'' \ 300 ppm POUR PLHIGH B.P. MATL WITH300 ppm POUR

\"'\ \ DEPRESSANT

\\ \

l-=J-60 -55 -50

TEMPERATURE, °C-45

Figure 6. - Fuel freezing behavior curves, comparing higher-boiling-point fuel blends with and without additive.

-lOr—

_o

UJ INSULATED TANK

0

Figure 7. - Predicted fuel temperature for a 9100 km com-mercial aircraft mission, with 2.5 cm-thick fuel tankinsulation, k = 0.65 W/lm-K).

o

LU~Di

o:LUQ_

20

10

0

-10

-20

-30

-40

-50

79 kW PER TANK FUELHEATING PLUS INSULATION-

79 kW PER TANK FUEL HEATING

20 kW PER TANK FUELHEATING PLUS INSULATION

— NO INSULA-TION OR HEATING

1 2 3 4 5 6 7 8TIME, hr

Figure 8. - Predicted fuel temperature for a 9100 km com-mercial aircraft mission, with heated and insulatedfuel tank.

o00oI

W

TANK BOOSTPUMP

*Ov_y

HEATED FUEL

ENGINE JDRIVEN(PUMP

3-WAYANDRELATING

r\

>

TANK

"2

RELIEFCIRCUVALV1

rFUELCONTROL

i

-E

1

^

iit

ADDED HEATEXCHANGER

i1

ii

l

LUBE OILCOOLER

1

1

HOT LUBEAll CD A ft/1UIL 1 KUM

ENGINE

1|

FUEL TO-^ENGINE

MANIFOLD

^.COOLEDLUBE OIL

WING TANK ENGINE-MOUNTED COMPONENTS

Figure 9. - Schematic of fuel heating system, using lubricating oil heatexchanger.

CJo

sQ-

100,—

80

I 60t—

x 40

-20

-30

-40

-50

FUEL HEATING SCHEDULE

NO HEATING

I I I ! !0 1 2 3 4

TIME,5 6 7

hr

18 9

Figure 10. - Predicted fuel temperature for a 9100 km com-mercial aircraft mission, with fuel heated by lubricatingoil heat rejection.

1

4

Report No.

NASA TM-79015Title and Subtitle

2. Government Accession No.

HIGH-FREEZING-POINT FUELS TEED FOR AVIATION

7

9

12

15

16

TURBINE ENGINES

Author(s)

Robert Friedman

Performing Organization Name and Address

National Aeronautics and SpaceLewis Research Center(""IpvplanH Ohin 441 Ti

Sponsoring Agency Name and Address

National Aeronautics and SpaceWashington, B.C. 20546

Supplementary Notes

Administration

3. Recipient's Catalog No.

5. Report Date

6. Performing Organization Code

8. Performing Organization Report No.

E-980410. Work Uni t No.

11. Contract or Grant No.

13. Type of Report and Period Covered

Technical Memorandum14. Sponsoring Agency Code

Abstract

Broadened-specification aviation fuels could be produced from a greater fraction of crude sourcematerial with improvements in fuel supply and price. These fuels, particularly those with in-creased final boiling temperatures, would have higher freezing temperatures than current avia-tion turbine fuels. The higher -freezing -point fuels can be substituted in the majority of presentcommercial flights, since temperature data indicate that in-flight fuel temperatures are relative-ly mild. For the small but significant fraction of commercial flights where low fuel temperaturesmake higher freezing-point fuelmade to accommodate this fuel.

use unacceptable, adaptations to the fuel or fuel system may beSeveral techniques are discussed. Fuel heating is the most

promising concept. One simple system design uses existing heat rejection from the fuel-lubricating oil cooler, another uses an engine -driven generator for electrical heating. Both sys-tems offer advantages that outweigh the obvious

17.

19.

Key Words (Suggested by Author(s ) )

Jet fuels; Hydrocarbon fuels; Aircraft fuelsystems; Turbine fuels; Freezing point

Security Classif. (of this report)

Unclassified

penalties.

18. Distribution Statement

Unclassified - unlimitedSTAR Category 28

20. Security Classif. (of this pagel

Unclassified

21. No. of Pages 22. Price"

' For sale by the National Technical Information Service, Springfield. Virginia 22161

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

WASHINGTON. D.C. 2O546

OFFICIAL BUSINESS

PENALTY FOR PRIVATE USE »3OO SPECIAL FOURTH-CLASS RATEBOOK

POSTAGE AND FEES PAIDNATIONAL AERONAUTICS AND

SPACE ADMINISTRATION451

POSTMASTER : If Undellverable (Section 158Postal Manual) Do Not Return