ICASPAPERunNO. c's

HYPERSONIC TRANSPORTS - ECONOMICS AND ENVIRONMENTAL EFFECTS

byRichard H. Petersen, Chief, and Mark H. Waters, Aerospace Engineer

Aeronautical Missions and Technology Branch,NASA Advanced Concepts and Missions Division, OAST,

Moffett Field, California USA

TheEighthCongressOfthe

InternationalcounciloftheAeronauticalsciences

INTERNATIONAALCONGRESCENTRUMRAI-AMSTERDAM,THE NETHERLANDS AUGUST28 TO SEPTEMBER2, 1972

Price: 3. Dfl.

HYPERSONICTRANSPORTS - ECONOMICSAND ENVIRONMENTALEFFECTS

R. H. Petersen*and M. H. Waters**NASA Advanced Concepts and Missions Division, OAST

Moffett Field, California94035

Abstract

An economic analysis of hypersonic transportsis presentedto show projectedoperating costs(direct and indirect)and return on investment.Importantassumptionsare varied to determine theprobable range of values for operatingcosts andreturn on investment.

The environmentaleffects of hypersonic trans-ports are discussed and compared to current super-sonic transports. Estimates of sideline and fly-over noise are made for a typical hypersonic trans-port, and the sonic boom problem is analyzed anddiscussed. Since the exhaust products from liquidhydrogen-fueledengines differ from those of kero-sene-fueledaircraft, a qualitativeassessment ofair pollutioneffects is made.

Introduction

Over the past ten years there have beennumerous studies of hypersonic transportsfor com-mercial application. References 1-10 give examplesof studies performed in the United States. Primar-ily these studies have concentratedon the perfor-mance aspects of hypersonic aircraft.

In the last few years, increasingemphasis hasbeen placed on the economic evaluationof proposedcommercialaircraft and on early study of their en-vironmentalcharacteristics. The purpose of thispaper is to provide current estimatesof the eco-nomics and environmentaleffects of hypersonic air-craft. Since the introductionof such aircraft isnot likely before 1990, or perhaps the year 2000,the estimates given in this paper can only be con-sidered crude approximations. Hopefully, they willprovide useful insight into the probablecharacter-istics of hypersonic commercial transportsand theproblems to be surmounted during their development.

To put the results in proper context, the papercommenceswith a brief review of aircraft character-istics and performance,followed by a discussion ofthe prime technologicalproblemsassociatedwithhypersonicaircraft. Then estimatesof economicperformanceare presented and discussed in detail.Finally, the characteristicswhich affect theenvironmentare analyzed to determinepotentialenvironmentalproblems associatedwith hypersonicaircraft.

ConfigurationsandPerformance

The nominal airplane considered in this studyis an all-body hypersonic transportconfigurationwith a gross weight of 1 million lb (454,000 kg),a cruise speed of Mach 6, and a range of 5500 n. mi.(10,200 km). It is accelerated to Mach 3.5 byturbojetengines and cruises at approximately100,000 ft (31 km) altitude on ramjet erminesfueled with liquid hydrogen. Such an aircraftcould carry 400 passengers between Los Angelesand Amsterdam in less than 2.5 hours.

A number of configurationoptions have beenexamined over the years. Figure 1 shows an all-body configuration,and figure 2 shows a wing-body.In general the all-body shape is well suited toramjet engines, particularlythose which employsupersonic combustion,because its body surfacescan be utilized as inlet and expansion nozzle.(Supersoniccombustion ramjetswould allow cruisespeeds of Mach 8 to 10.) On the other hand, theall-body has high drag characteristicsat transonicspeeds where the acceleratorengines are sized.The all-body has better structuralweight character-istics, but the wing-body has superior aerodynamicperformance,particularlyat transonic speeds.Both of the configurationoptions have.similar per-formance in the speed range from Mach 6 to 8 wheretheir performance is optimum;a Mach 6 all-bodyaircraft using subsonic burning ramjets was chosenas representativefor this study.

Figure 1. All-Body Configuration

Figure 2. Wing-BodyConfiguration

Another option for hypersonic transportationis the rocket-poweredboost-glider. There are manyproponents of boost-glidehypersonic transports,and their analyses make the boost-glider appearpromising, but these studies normally assume verylow structuralweights. The results from a recentunpublished study which compared boost-glide ve-hicles with rocket-boosted,airbreathing•cruisevehicles and with all-airbreathingvehicles arepresented in figures 3, 4, and 5. In this study,all three vehicles were analyzed using the sameweight estimating techniques,and the all-airbreath-ing transport clearly gave the best economic per-formance. Note that the direct operating costs dueto propellants for the rocket-boostedvehicles were

*Chief, AeronauticalMissions and Technology Branch.**AerospaceEngineer, AeronauticalMissions and Technology Branch.

1

considerablyhigher than the total direct operatingcosts for the airbreathingvehicle. With the addedproblems of passenger acceptance(due to relativelyhigh accelerationsand periods of weightlessness),the boost-glidemode of transportationdoes notappear competitive with a hypersonicairbreathingaircraft.

Figure 3. Weight Breakdownfor Rocket Boosted andAirbreathingHypersonicTransports

Figure 4. Acquisition Costs for Rocket Boostedand AirbreathingHypersonic Transports

BOOST GL DE ROCKET RJ/SJ AIM:WEATHER(TWO STAGE)

Figure 5. Operating Cost Breakdown for RocketBoosted and AirbreathingHypersonicTransports

TechnologyStatus

While there will be many aerodynamicproblemsassociatedwith the developmentof a hypersonictransport, aerodynamicsdoes not appear to be apacing technology. Wing-body configurations,all-body configurations,and variationsbetweenthese two configurations(which are often calledblended bodies) have been under study for severalyears. Probably the biggest aerodynamicconcern isthe efficient integrationof the propulsionsysteminto the overall configuration. Developmentandoperation of the space shuttle vehicle in thenext decade will provide useful aerodynamicexper-ience for hypersonictransport designers.

There are several design approaches to tbestructure and thermal protectionsystem of a hyper-sonic transport. Because of the low density ofliquid hydrogen, it is not feasible to carry muchfuel in the wings of a wing-body configuration.Typically, fuel is carried in the fuselage andeither integral or nonintegraltanks may be used.With integral tanks the tank structure carriesthe primary bending loads in the fuselage;non-integral tanks are separate structuresset withina load-bearingfuselagestructure. In general,the integral tankage is more efficient but is amore difficult design concept. Because of its non-circular cross section, the all-body shape doesnot lend itself to nonintegraltankage and normallyuses integral tankage consisting of a number ofconical, intersecting,tank sections. Again thisis a difficult design problem, but analyticalstudies show this to be a highly efficientstructure.11

The structuraldesigner has the option ofusing high temperaturematerials for the structure,to resist the heat generated at hypersonicspeeds,or employing normal aircraft materials and usinga thermal protectionsystem to maintain low tem-peratures in the primary structure of the vehicle.Preliminaryanalyses indicate that these twoapproaches may be competitiveup to Mach numbersof 5 or 6, but at higher speeds the hot structureconcept becomes much heavier than the cool structureconcept. The conclusionof most recent studies isthat an integral structure of aluminum and/or tita-nium, combined with a suitable thermal protectionsystem, is the best approach for a hypersonictransport. Obviously,an aluminum structure ishighly desirable in terms of meeting the long liferequirementsfor a commercialairplane at reason-able costs.

The best design of a suitable thermal protec-tion system presently is not clear. Basically,there are three elements which can be used to designa thermal protectionsystem. These are activecooling, in which a liquid or gas coolant is circu-lated through the surface and conducts heat awayfrom the surface, insulation,which may be usedto restrict the flow of heat into the structure,and radiation shields,which may be used to en-courage radiation of heat away from the surface.

The three elements, active cooling, insula-tion and radiationshielding,can be combinedin a number of ways. One promising approach8 isto use an actively-cooledstructurewith secondarycoolant circulatingthrough tubes beneath thesurface of the structure and carrying the heatto the hydrogen fuel. With careful design of

BOOST GLIDE ROCKET RJ/SJ

M.6

198 PASSENGERS

RANGE.5500 11. mi.E3 OXIDIZER (L02)

C:3 FUEL (LH2)

C:3 OPERATING WEIGHT EMPTY

=I) PAYLOAD

2.0

SECOND STAGE

FIRST

STAGE

ARBREATHER M.8

198 PASSENGERS

250 AIRCRAFT FLEETRANGE.5500 n. mi.

SECOND STAGE

BOOSTER

BOOST GLIDE ROCKE T RJ/SJ AIRBREATHER

(TWO STAGE)

EZI FUELCD OXIDIZER=I MAINTENANCEC=I DEPRECIATIONEzt CREW +INSURANCE

15 PBSPASSENGERSRANGE.5500

LIII

7.5

§ 5.0

u.2.5Li

2

the airframe and engine, it appears that the entireairframe and the cruise engine could be cooled withthe hydrogen required for combustionin the engine,for flight Mach numbers up to 8. In areas wherethe heat transfer is high, radiationshields canbe used to significantlyreduce the heat transferredto the coolant. Another promisingapproach7 usesinsulationand radiation shields to maintain thestructure at a low temperature,and the fuel isused to cool only the cruise engines. This alsoappears to be a practical approach. In both cases,further insulation is required between the surfaceof the hydrogen tank and the structure,which oper-ates near room temperature. Air must be excludedfrom this insulationto prevent cryopumpingcausedby air liquifaction.

The space shuttle will provide valuable experi-ence with structuresand thermal protectionsystemsat hypersonic speeds, but it experienceshigh heatloads for short times while the hypersonic transportwill see moderate heat loads for long times. Also,the hypersonic transport must be designed for amuch greater structural lifetime. As a result,the transportwill use differentconcepts and con-siderable research on these concepts is required.

The pacing item for hypersonic transportdevelopmentis propulsion system technology. Theoperation of small-scale subsonic and supersonicburning ramjets has been adequatelydemonstrated.On the other hand, there are difficultproblemswith the fabricationof the ramjetsbecause theirsurfaces must be regenerativelycooled with liquidhydrogen fuel as it flows to the combustor. TheNASA Hypersonic Research Engine,which is nowin test, will provide some experiencewith regen-erative cooling for an airbreathingengine.

Propulsionsystem researchmust also befocused on designs which are integratedwiththe overall aircraft configuration,in contrastto previousexperimental ramjetswhich have beenaxisymmetric. Most recent study configurationshave nonaxisymmetricramjetswith the acceleratorturbojets located in a separate duct.12 (Theinlet may or may not be common to both engines.)

At present the risks associatedwith hypersonictransportdevelopment are very high because of theinadequateresearch experiencewith their propulsionsystems. The necessary confidencecan only be ob-tained through testing of larger engines and engineswhich are integrated into practicalaircraft de-signs. This research will be costly and time con-suming and is hampered by the current lack of largescale test facilities. Because of the lack ofpropulsiontechnology the expected operationaldate for hypersonic transports is beyond 1990,and unfortunately,the space shuttle developmentwill provide little technology applicableto air-breathing hypersonic engines.

Economic Evaluation

As indicated previously, the nominal vehiclefor the economic studies was a 1 million lb (454,000kg), all-body,Mach 6 hypersonic transport. Thedesign of this transport was optimized to obtainbest performancewith the all-body shape.7 Otherconfigurationand engine combinationsmight provideimproved performance,but it is felt that this con-figuration is representative. The aircraft per-formancewas estimated using a synthesis program

developed by NASA's Advanced Concepts and MissionsDivision over a ten-year period. This programis described in reference 7.

Operating costs were estimated using standardmethods of the U. S. Air TransportAssociation andU. S. manufacturers with appropriateadjustmentsfor hypersonic aircraft. Aircraft development andproduction cost estimateswere based on aircraftweight, complexity, and speed capability. All costsare in 1972 U. S. dollars. The nominal case makesthe following assumptions: (1) hydrogen costis 10 cents per lb (22 cents per kg), (2) turn-around time is 1.5 hours, (3) fraction of dayin use is 0.5, (4) passengerload factor is 50%,

number of aircraft produced is 250, andreserve fuel is 5% of block fuel plus 45-minute

hold. Current internationalfares were used tocompute revenue. Aircraft prices were determinedby estimating developmentand production costsfor a given fleet size (nominalequals 250), addinga 10% profit for the manufacturer,and dividingby the fleet size.

Cash flow return on investment (ROI) is deter-mined by assuming an airline investment equalto the cost of the aircraft,plus 10% airframespares and 40% engine spares, and estimating theyearly cash flow as the differencebetween yearlyrevenues and operating costs (not including depreci-ation). The ROI is then the yearly cash flowdivided by the investment. ROI is a better measureof the economic value of an aircraft than directoperating cost (DOC) because DOC does not takeinto account the productivityof the airplane.An airplane with a higher DOC will not produceas much revenue per passenger-mile,but if itis much more productive (in terms of seat-milesper year), it can easily generate just as muchROI. The hypersonic transportunder study herewould generate about 2.3 billion seat-miles peryear compared to 0.5 billion seat-miles per yearfor the Concorde and 1.0 billion seat-miles peryear for the Boeing 747.

The basic economic performanceof the nominalhypersonic transport is shown in figure 6. Threevehicleswith different design ranges are indicated.All have a gross weight of 1 million lb; the 5500-n. mi. (10,200-km)aircraft carries 404 passengers,the 4500-n. mi. (8300-km)aircraft carries 540passengers,and the 3500-n.mi. (6500-km)aircraftcarries 684 passengers. The direct operatingcost for the 5500-n. mi. design varies between2 and 2.5 cents per seat-mile for ranges between3000 and 5500 n. mi. The 4500- and 3500-n. ml.designs have better economic performance. The

2.0

o 1.0

Li

ci 0

4

r E

in 2

:t)0 0

0

80-

-DESIGN RANGE t

n. mi. 1560a

-55004500 0

3500m 40

0

4" 20 -35004500

-5500

\

2 46 02

RANGE, n.m x1034

4 8 12 C4 8

12

RANGE, knl ( I03

Figure 6. Effect of Range on HST Economics

3

DOC of a 3500-n. mi. design at its design rangewould be slightly more than 1.5 cents per seat-mile which is comparable to the Concorde. TheDOC of a 747 is in the neighborhoodof 1 centper seat-mile. On the other hand, the 5500-n. mi.vehicle produces an ROI between 10% and 20% forranges of 3000 to 9500 n. mi. A typical routestructuremight result in an overall ROI of 15%,which is marginal but should be acceptable. (Notethat this return assumes current fares and a 50%load factor.) The vehiclesdesigned for shorterranges obviously provide a more attractiveROI.

One might ask why a nominal design range of5500 n. mi. was chosen, although it clearly penal-izes the economic performanceof the vehicle.Figure 7 shows cumulative internationalair passen-gers versus range for 1980, and figure 8 illustratesthe time required to fly various distanceswithdifferent aircraft types. A design range of 3500n. mi. encompasses 50% of the world travel, buta design range of 5500 n. mi. would include about90%. Figure 8 shows a not unexpectedtrend; a Mach6 aircraft shows a modest time saving over theConcorde at ranges up to 3000 n. mi. At longerranges, the time saving and thereforethe competi-tive advantage of a hypersonictransportbecomeslarger, particularlyif the limited range of theConcorde forces a stopoveror change of airplanes.For these qualitative reasons,a design range of5500 n. mi. was selected for the nominal hypersonictransport.

2 4 6 8 10RANGE, 1000 nJni.

0 2 4 6 8 10 2 A 6 6

RANGE, 1000 km

Fi-ure 7. Cumulative InternationalPassengerTraffic with IncreasingRange -1980 Estimate

TO

2 3 4RANGE, nJTIL x 103

4 6 8

RANGE, km x Q4411,

- WEST TO U.S. - EAST TO EUROPE

US - WEST TO HiwvAu 10 INONSTOPI JAPAN 0.0A HAWAW

..S. - EAST TO EUROPE TO LEBANON

- EAST To U. S. - WEST TO HAWAII

Figure 8. Schedule Time Comparisonfor Subsonic,Supersonic,and Hypersonic Transports

To put the economic results in context, table 1 gives a weight breakdownof the nominal5500-n. mi. aircraft,table 2 summarizes the acqui-sition cost items, and table 3 gives a breakdownof direct and indirectoperating costs. Fromtable 3, it is clear that fuel forms a major por-tion of the operatingcosts for a hypersonictrans-port. The costs shown in the table are for liquidhydrogen at 10 cents per lb (22 cents per kg).A study done severalyears ago" indicatedthatthis was an achievablecost for liquid hydrogen inlarge quantities in the 1990's. For comparison,ata cost of 4.2 cents per lb (9.3 cents per kg),liquid hydrogenwould provide about the same energyoutput per dollar as current jet fuels. Althoughsome authors14 have projectedcosts as low as4 cents per lb, such costs do not appear likelywithin this century.

AIRFRAME STRUCTURE 40.4%PROPULSIONSYSTEM 10.1%FIXED EQUIPMENT 5.7%FUEL 35.3%PAYLOAD 8.5%

TABLE 1 WEIGHT BREAKDOWN

Millions of Dollars

RDT&E 6214.5AIRFRAMEDESIGN& DEVEL-OPMENTENGINEERING 2350.5

JISCELLANEOUSSUBSYSTEMDEVELOPMENT 116.8

PROPULSIONDEVELOPMENT 1790.9

DEVELOPMENTSUPPORT(INCLUDES5 FLIGHTTESTVEHICLES) 1956.3

INITIALINVESTMENT 9695.5OPERATIONALVEHICLES(245) 7882.9

SUSTAININGENGINEERINGANDTOOLING 1615.2

OTHER 197.4

PROFIT 1591.0

TOTAL 17,501

AIRCRAFTPRICE= $70x106

TABLE2ACQUISITIONCOSTBREAKDOWN

DIRECTOPERATINGCOST

Cents/Seat-Mile

2.040CREW 0.02

FUEL 1.40

INSURANCE 0.05

MAINTENANCE 0.37

DEPRECIATION 0.20

INDIRECTOPERATINGCOST

.907DIRECTMAINTENANCE .108

AIRCRAFTSERVICING .003

PASSENGERSERVICE .516

TRAFFICSERVICE .129

GENERAL& ADMINISTRATIVE .151

TABLE3 OPERATINGCOSTBREAKDOWN

GROUND RANELJGER •

30 min 03140PLANE CHANGE STOP •

20 Inn10 • O.'

PLANE CHANGE

6.0

10

8AU

2 7-6

- I

•

4

100

Lito- 80z cw

60

t 40

4 a

20

0

4

Figure 9 indicates the strong effect of fuelcost on the economics of the hypersonictransport.The high cost of fuel would be a severe problemduring the introductionof hypersonictransports.Using current technologiesand a reasonablylargevolume of production,liquid hydrogencould beproduced for about 15 cents per lb (33 cents perkg); this would be a reasonableestimate of the fuelcost early in the introductionof the hypersonicfleet. Figure 9 indicates that the ROI will bevery low with this fuel cost but will improveconsiderablyas the cost of liquid hydrogen isreduced.

2.0

'61.5

0

r. 1.oa,

75

0

4

3

0 2

LI

FUEL COST

Cents/lb

15

10

5

.

80

at60

-ci1rt 40

3 o

1- 20

a

.u 1

0

,,-v ---15 1

C

2

,

4 602

RANGE, n.rni.x103

4 6

4

81204 8 12

RANGE, km x103

Figure 9. Effect of Fuel Cost on HST Economics

Fuel reserve requirementsalso have a stronginfluenceon economic performanceas shown infigure 10. The nominal case assumed reservesequal to 5% of the block fuel, plus sufficient fuelto hold 45 minutes in subsonic flight. As indi-cated in the figure, an increase from 5% of blockfuel to 10% of block fuel results in a drop inROI of about 5%, which is very significant. Theeffect is not due to the cost of reserve fuel(in fact, only block fuel is included in oper-ating costs), but is due to the reductioninpayload in order to carry more reserve fuel. (Theaircraftwith 10% reserves carries 361 passengersas compared to the nominal aircraftwhich carries404 passengers.) Obviously there will be a sig-nificant economic payoff if the reserve fuel re-quirementcan be reduced. Such reductionsshouldbe possiblewith the highly automatedair trafficcontrol systems expected in the 1990's and therelativelyshort flight times of hypersonicair-craft.

E 2.0

4

A

.3

80

- RESERVES,FRACTIONOf BLOCKFUEL•

0.4160

0.10

1.5

.05

0 20

ci40

0

I

20

.5

0.0105

0

2 460 2 4 6

RANGE, n.mi.x103L_J L1_ I

04 8120

4

12EMOTIONLNEI 1 memo 0.

01.11111ESWSW( CIWSE DESCENTRANGE, km x103

Figure 10. Effect of Fuel Reserve Requirementson HST Economics

The next two figures, figure 11 and 12, relateto the utilization of aircraft. Figure 11 showsthe effect of turnaroundtime on economic perfor-mance. It is difficult to reduce the turnaroundtime of a hypersonic transportbecause of the ther-mal problems involved in loading the cryogenic fuel,particularlyif a flight has just been completedand the vehicle is still warm. The nominal caseassumed a turnaround time of 1.5 hours, and figure11 indicates that turnaroundtime did not have alarge effect on economic performance. On the otherhand, a 2% increase in ROI, which is achievable byreducing turnaround time to 1 hour, means a verylarge financial return to the operator.

Figure 11. Effect of Turnaround Time on HSTEconomics

Figure 12 shows the effect of aircraft timeon line. The nominal case assumed a factor of 0.5,meaning that, on the average,each aircraft is oper-ated for 12 hours a day. This time includes bothblock time and turnaroundtime so that flight hoursare considerably lower (for the 5500-n. mi. nominalcase the average flight hours were 6.8 hours perday). As in the case of turnaroundtime, thefraction of time on line does not have a strongeffect on economics but the operator will seelarge returns from increasinghis time on linefactor from 50% to 60%.

4

2.03

i51.5 15,0

0 2

N 1.0

aJ 5Li I

2460246

RANGE, n. rm. x10 3

4 8 12 0 4 8 12

RANGE, km x 103

Figure 12. Effect of Time on Line on HST Economics

All of the results shown to date have beenbased on normal internationalfare structures anda 50% passenger load factor. It is reasonable tbexpect that a Mach 6 transportoperating at today'sfare levels will be very attractive to the air

4 80

4 66 0 2

RANGE, n. mi.x 103

0

1 I

40 2

1.0E3

0 2

1160

•

4 8 12 0 4 8 12

RANGE, km x103

E 2.0

1.5

Lo

a0

ce 40

3

u-

20

an

TURNAROUNDTINE,hr

2.01.5

1.0

0.4.5.6

80

60a

ci 40

3 0

200 a

FRAC:ICIPEOf TIME.01Tif(ROCKANCTUANAVSCPC'

0.6.5

5

traveler. Figure 13 indicatesthe effect ofload factor on the economic return. At a 60%load factor, which does not seem unreasonable,the ROI ranges from 15% to 30% dependingon rangeand should be well over 20% for a typical routestructure. At a 70% load factor, the return couldbe as high as 30%, but such load factors are gener-ally considered unattractivebecause too manypassengersare unable to get the flight they want.Clearly, if the hypersonictransportcan draw 60%load factors, its economics look very promising.

4

80

2.03

k60

1.5

2

ce 40

01.0

5,....5 71

u- 20

6

o 0

0 2 460

RANGE, n.mi.x 103

0

,

4 8120

RANGE, km x103

Figure 13. Effect of PassengerLoad Factor onHST Economics

Another method of increasingthe ROI is toplace a surcharge on hypersonictransport fares.The effects of such surchargesare shown in figure14. If a an; surcharge can be applied while stillmaintaininga 50% load factor, the effect is aboutthe same as a 60% load factorwith no surcharge;i.e., the ROI is better than 20%. Note that all ofthe curves on figure 14 are for mixed seating with20% first class and 80% coach. If the aircraftwere designed with all first class seating andcurrent internationalfirst class fares were chargedto all passengers, the ROI would fall very close tothe 40% surcharge line shown on the figure. TheresultingROI of better than 30% makes this a veryattractiveoption.

8 4 8 12 0 4i 1 t 1

8

RANGE, km x 103

Figure 14. Effect of Fare on HST Economics

As indicated in table 2 the aircraft priceassumed for the nominal case was $70 million.

Estimatesof this price are certainlyopen to ques-tion, and figure 15 illustratesthe sensitivityofeconomic performance to increasesin aircraft price.An increase of 50% in aircraft price, to $105

million, would reduce the ROI by about 5%. If onlyincreases in developmentcost were considered,a50% increase in developmentcost would resultin a 20% increase in the price of the aircraft.It should be noted again that all prices are quotedin 1972 U. S. dollars and do not include increasesdue to inflation.

2.0

I.5

e 1.0

r

0

4

3

,`L

0 2

LiI

0

80-

R60

ss,4020

a—X40

20

PERCENTINCREASEIN NOMINALkiCRAFT ACQUISITIONCOST

020

40

2 460246

RANGE , n. mi. x IC 3

8120412

RANGE, km x 103

Figure 15. Effectof AcquisitionCost on HST Economics

Aircraft price is also strongly influencedbythe number of aircraftwhich the manufacturerex-pects to sell. The nominal fleet size consideredhere is 250, and figure 16 indicatesthe effect offleet sizes of 150 and 350. With a fleet sizeof 150 the aircraft price is about $100 million;with a fleet of 350 it is about $64 million. Thelower fleet size resultsin about a 4% reduction inROI, and the larger fleet gives about a 3% increase.

4 - NUMBER OFAIRCRAFT

g

x20 - 350 -------

0 _ 250.,\,,..../5%,a 150--o

2 4 6 0 2 4

RANGE, n.m..x 103, I , I I I

I I84 8 12 0 4 ' 112

RANGE, km x I03

Figure 16. Effect of Fleet Size on HST Economics

A natural question is, "How big a market forhypersonic transportationwill there be in the1990's?" Some indicationis given in figure 17.The upper curve shows the projected growth of totalinternationalair transportation. Two estimatesof long range travel are also shown. The middlecurve is a Boeing estimate of revenue passengermiles for SST overwaterflights, and the circle"is an estimate of revenuepassenger miles forall internationalroutes at ranges greater than3000 n. mi. in 1990. Also shown in the figureis a conservativeestimateof the hypersonictrans-port market. This curve is based on the assumptionthat the supersonicfleet in operation in 1990continues to operate;otherwise the hypersonicestimate would be much larger. About 300 aircraft

LOAD FACTOR

4 8 12

MIXEDSEATING

iRsT CLASS 20%

COACH 80%

80

k60

ce40

20

0

PERCENTINCREASEIN INTERNATIONALFARE

24 602

RANGE, n.mi.x 103

2.0

1.5

0 1.0

Li

d 0

80

,$) L,

k 60

ce 40

6

6

would be required to service the market shown inthe year 2000, and this number would grow rapidlyif productionwas extended another five or tenyears.

Before summarizing the economic conclusions, attention should be focused on the extremely large capital investmentwhich would be required to devel-

op and produce a hypersonic transport. Table 2 indicatesa development cost of $6.2 billion, but the problem is more clearly illustratedby examin-

ing figure 18, which shows the estimated cash flow of the manufacturer for a program in which 250 air-

craft are produced in a ten-yearperiod. Note that the peak investment in the program is approximately $6 billion, about 13 years after start of develop-

ment, and the cumulative cash flow does not become positive until 17 years from the start of develop-

ment. It is clear that no single aircraft manu-

facturerwill be able to undertakea developmentof this size, and few governmentscould support suchan effort unilaterally. There must either be a verylarge cooperativeeffort involvinggovernmentsandindustry,or the hypersonic transportwill not be-come a reality.

Figure 18. Cash Flow Breakdownfor 250Aircraft Fleet

In this study, aircraft price was determinedby totaling development and productioncosts andadding a 10% profit. Because of the long timeperiod between investment and return, the discountedcash flow return on investmentto the manufactureris only 3% for the program shown in figure 18.If the production run was extended six years anda total of 500 aircraft were manufacturedand soldfor $70 million each, the cash flow ROI would beabout 10%. Alternately, the cash flow ROI fromthe program shown in figure 18 could be raised to

10% by raising the aircraftprice to $93 million.The effect of this 33% price increase on airlineROI can be seen in figure 15.

In summary, the resultsof this very preliminarystudy indicate that the nominal hypersonic transportconsidered here is marginalon an economical basis.Designing for ranges shorter than 5500 n. mi.(10,200 km) dramaticallyimproves the economicsbut reduces the potentialof the aircraft forlong range transportation. The cost of liquidhydrogen is crucial to the economics of the air-plane, and costs of 10 cents per lb (22 cents perkg) are desirable. Moderate but significantimprovements in economic performancecan be obtainedby minimizing fuel reserve requirements,reducingturnaround time, and increasingaircraft timeon line. If a 60% passengerload factor canbe achieved with current internationalfaresrather than the 50% assumed in the nominal case,hypersonic transporteconomics look promising.Alternately, if a 50% load factor could be achievedon an all first class airplane at first classfares, the economics look very good. Finally,because of the relativelysmall fleet size andvery large capital investmentrequirements,onlya large cooperativeeffort involving aerospacemanufacturersand governmentswill brina aboutthe development of a hypersonictransport.

EnvironmentalConsiderations

It is important that the potential effect onthe environment be consideredin the design of anynew aircraft system. This is particularly true forsupersonic or hypersonictransport aircraft becausethese aircraft cruise in the stratospherewhere theresidence time of the engine exhaust products willbe measured in years rather than days or hours asis the case for aircraft cruising in the tr000-sphere.

The question of environmentaleffects had aprofound effect on the decision to cancel theAmerican SST. The areas of concern involvedthe noise generationduring takeoff, the potentialadverse effect of exhaust products on the strato-sphere, and the sonic boom overpressure at super-sonic flight speeds. A hypersonic transportwill create the same concerns as the SST, andin this section the object is to present a briefdiscussion of the Potentialproblem areas andto assess the magnitude of each.

Takeoff NoiseAn airbreathing hypersonictransport will

cruise with ramjet engines, but separate acceler-ator engines must be provided for takeoff andacceleration up to approximatelyMach number 3.5.These acceleratorengines also burn liquid hydrogenfuel, and the stoichiometricburning turbojetis a promising concept. With stoichiometricburningin the combustor there is no afterburner,butthe modest cycle pressure ratio required for super-sonic flight (10-15) leads to a very high exhaustvelocity at rated power. The result is a highlevel of jet noise. The noise from the rotatingmachinery will be negligibleby comparison, andis ignored in the noise estimates which follow.

A mitigating factor with regard to jet noiseis the capability to take off with engines throt-tled. The acceleratorengines for both the wing-body hypersonic transportand the all-body hyper-

BOEING ESTIMATE-

SUPERSONIC OVER WATER

HYPERSONIC ESTIMATE

01960 1970 1980 1990 2000

TIME, yr

Figure 17. Revenue PassengerMile Projectionsfor InternationalTraffic

coo -50 2 TOTAL

wrc - 800INTERNATIONAL

[

0 ESTIMATE OF RPM FOR INTERNATIONAL

ROUTES GREATER THAN 3000 n.ml.

4000-

MOO

cT 2000

-I,

•

1000 REID

0

6 6000

4000

g

2000

-I 0

2L 2000

-4000

-60000 5 10 15 20

YEARS AFTER PROJECT MITIALIZATION

/•,II

-11cSALES

11

--s-MANUFACTURING

CUMULATIVE CASH FLOW

7

sonic transport are sized to provide adequatethrust to accelerate transonically. In the caseof the wing-body, this yields a ratio of takeoffthrust to gross weight of about 0.4,2 and for theall-body this ratio is about 0.85.7

Figures 19 and 20 show noise levels, takeoffdistance, field length, and initial ground rollacceleration for a 1 million lb (454,000 kg) hyper-sonic transport at takeoff thrust to gross weightratios of 0.4 and 0.85. A maximum lift coefficientof 1.0 is assumed, and the maximumnormal loadfactor is limited to 1.2. As shown in figure19, noise levels at full rated power are veryhigh--131 to 136 PNdB at the sideline measuringpoint and 120 PNdB at the downrange measuringpoint. As the engines are throttled to a lowerthrust level and thus lower jet velocity, the noiselevel drops rapidly, and at 60% of full power thenoise is down to about 80 PNdB. This estimate maybe slightly low because the jet noise theory issuspect at low jet velocities and because othernoise sources may become audible, but the noiseis certainly below 90 PNdB. Basically the largesize of the accelerator engines allows just theproper thing, movement of a large amount of airslowly to obtain thrust.

MAXIMUM LIFT COEFFICIENT . 1.0

MAXIMUM SEA LEVEL THRUST

GROSS WEIGHT

Figure 19. Estimate of Sideline and FlyoverNoise During Takeoff

MAXIMUM LIFT COEFFICIENT .1.0

Thipx/W - MAXIMUM SEA LEVEL THRUST/GROSS WEIGHT

Field length requirements are shown in figure20. Field length is equal to the largest of threelengths: takeoff distance times 1.15, balancedfield lenath, and landing distance divided by0.6. The all-body, with its very high thrust,can operate from 4000-ft (1200-m) runways evenwhen throttled to 60% power; the wing-body wouldneed 7000 ft (2100 m) to take off at 60% power.Looking at the acceleration chart on figure 20,it appears that the all-body would have to bethrottled to 60% or perhaps 50% on takeoff fromthe standpoint of passenger comfort.

The conclusion to be drawn from these estimatesis that the high maximumthrust loading of a hyper-sonic transport will allow these aircraft to takeoff with engines throttled, significantly betterthe requirements of current noise regulations*,and still operate from field lengths less than10,000 ft (3000 m). Additional calculationsof EPNdB, which corrects the PNdB noise level toaccount for both the subjective response to toneand a duration factor, indicate that the EPNeBlevel may be 5 to 7 dB less than the PNdBdatashown in figure 19.

Atmospheric Pollution rolliarein of the atmosphere due to engine

exhaust products can be broadly classified intotwo regions: the airport vicinity during groundmaneuvering and takeoff and the stratosphere duringcruise. Current problems in the vicinity of theairport stem primarily from the emission of un-burned hydrocarbons and carbon monoxide; the useof liquid hydrogen fuel will eliminate these prob-lem due to the absence of carbon in the fuel.Emission of the oxides of nitrogen (N0r) is also aproblem due to photochemical reactions that createsmog. This problem is prevalent for engines withhigh cycle pressure ratios, 15,16 and may be rela-tively minor for hypersonic aircraft because theirengine cycle pressure ratio is relatively low(half that of the current airbus engines). Alsothe hypersonic transport probably will take offwith throttled engines, lowering the engine cyclepressure even further. Thus it appears that atmo-spheric pollution in the airport vicinity willbe minor for liquid hydrogen fueled HST's, andthe discussion which follows considers strato-spheric pollution only.

At the time the American SST program was can-celled, there were several conflicting theoriesand much confusion about the potential pollution ofthe stratosphere and the resulting effect on theEarth's climate. Since that time, investigationshave progressed to the point where the potentialproblem are fairly well identified,17," but themagnitude of the pollution created by transportsflying in the stratosphere as related to naturalphenomena has not been determined.

Current studies are being done parametricallyto evaluate the effect of changing concentrationsof a particular substance on the total atmosphericheat balance. However, the atmosphere is extremelydynamic and involves horizontal and vertical trans-port in the troposphere with residence time measured

2 140-

E

120 -.6

100

La 80

LJ

60 '.6 .8 1.0 6 .8 1.0

THRUST/ MAXIMUM THRUST

0.85.40

•140

ID 120fn z

a

"7((.4±' 1010

2 ci.g 80

0.40

.85

TmAx/W.0.85

N

0 4 ....

il 1 - .4

4; 4r OAOcr, o8 P w ' 10

aOM i 2

I I-0 - 0 i 0L 0 '

.6 .8 LO .6 .8 1.0

THRUST/MAXRAUM THRUST

Figure 20. Estimate of Field Length andTakeoff Acceleration

100E-.- LANDING 12 - 40

— BALANCEDFIELD 04

[

LENGTH

O 1 "0 75 L- - TAKEOFF

i 0

_ , _ x 2 r x

30

--SMAX/W.0.4 o 8E

*FAR Part 36 requires maximum noise levels of 108 EPNO at both the sideline and downrange measuringstations.

8

in days, primarily horizontal transportin thestratospherewith residence time measured in years,and transfer between the troposphereand strato-sphere at the lower latitudesnear the equator."Accurate quantitativeeffects can only be deter-mined by accounting for the major dynamic inter-actions of atmosphericsubstances,but this isimpossiblewith present computer technology. Newcomputer developmentsmay make this dynamic modelingpossible, but to account for long range effectsover a period of years seems a remote possibil:ty.

As mentioned before, the exhaust productsof an HST will be free from moleculescontainingcarbon, and the potentiallyserious problems ofincreasedconcentrationsof carbon dioxide andunburned hydrocarbonparticulatesin the atmosphereare not factors. Problems that remain are thepotential increase in the concentrationof bothwater vapor and the oxides of nitrogen in thestratosphere.

Added water vapor in the stratosphere results in several effects on the atmosphericbalance of heat. The potential increaseof cloudsand contrails in the stratospherecan affect thereflectionand absorption of both ultravioletradiation from the sun and infrared radiationemitted from the Earth.17,20 For clouds to form,the ambient temperaturemust be below the frostpoint which decreases with altitude. Also, the tem-perature below which contrailswill form decreaseswith altitude. At the cruise altitudeof an HST(above 100,000 ft), indicationsare that theambient temperaturewill be above that necessaryfor either cloud or contrail formation.21

at present are uncertainand somewhat contradic-tory." The recent study by Ashby, et a1,22 dis-tinguished between the reactionstaking place indaytire or nighttime. At night much less regenera-tion of nitric oxide is predicted. Added to theproblem is the lack of any estimate of the naturalflux of NOx into the stratosphere.

As mentioned at the outset of this section,the potential problems of stratosphericpollutionare fairly well defined, but no definitive conclu-sions can yet be drawn concerning the magnitude ofthe problem, particularlywith respect to oxides ofnitrogen. Figure 21 presentsestimated data forthe production of both water vapor and nitricoxide per 1000 lbs of fuel for a JP fueled SSTand a liquid hydrogen fueled HST. The estimateof NOx for the SST was taken from reference 15 witha correction for the combustor pressure at altitude,and the SST water vapor estimate is for the completecombustion of C0H20-type fuel. The HST estimatesare from unpublishedNASA data for a fixed geometryscramjet module which is described in reference 27.Data are presented for the flow in chemical equili-brium at both the combustorexit and the nozzleexit. If the chemical kinetics in the nozzle resultin exhaust products which are not in chemicalequilibrium, then the productionof NOx can berelatively high, as shown at the combustor exit.On the other hand, the flow at the nozzle exit infigure 21 is not fully expanded, and if the flowcontinues to expand at equilibrium,the temperaturewill drop to a point where the NOx productionwill be negligible. The water vapor productionusing liquid hydrogen fuel is high. about 9 lbsper pound of fuel for complete combustion.

In addition, increased amounts of water vapormay react with the free oxygen, 0, in the atmo-sphere, reducing the availabilityof 0 to reactwith 02 and form ozone, 03. Ozone is the primaryabsorber of solar ultravioletradiation,and itsdepletionwould increase the amount of radiationthat reaches the surface of the Earth. A recentstudy by Ashby, Shimazaki, and Weinman.':indicatesthat the water vapor emissions that might be expec-ted from an SST fleet will have no noticeable ef-fect on the atmospheric concentrationof ozone.

WATER VAPOR,

1000

100 NO.

SST ESTIMATEHST SCRAMATS

0 MACH NO 4

6

8

NOZZLE EXIT

0 COMBUSTOR EXIT•

The current flux of water vanor from thetroposphereto the stratospheredue to Hadley cellcirculation" has been estimated to be approximate-ly 11,000 lb/sec (5 x 10 kg/sec).21 Anotherrecent estimate of the total flux of nlatervaporinto the stratosphere is 78,000 lb/sec (36 x 10'kg/sec).24 This estimate includes the additionalwater vapor from severe local thunderstorms,whichaccount for approximately56,000 lb/sec (25.4 x 10'kg/sec), and from natural oxidationof methane.Manabe and Wetherald2° have estimated that a five-fold increase in stratospherewater vapor concen-tration from the present value of 3 ppm to 15 ppmwould result in an increase of about 3.6°F (2°C)at the surface of the Earth and a decrease at analtitude of 65,000 ft (20 km) of 12.6°F (7°C).

The effect of increasing the concentrationofoxides of nitrogen (primarilyNO) in the strato-sphere is a subject of current debate. Separateinvestigationsby Johnston25 and Crutzen26 indicatethat nitric oxide may react with ozone in a complexway that ends with nitric-oxide being produced ina self-regenerativefashion along with 02. Results

10 ,1000 2000 3000 4000 5000

°R

_1 _L

1000 2000 3000GAS TEMPERATURE, •K

Figure 21. Estimate of Water Vapor and NOxEmissionsDuring Cruise

To make an estimate of the annual productionof water vapor and NOx by an HST fleet, data forthe Mach 6 scramjet with equilibrium flow at thenozzle exit was used from figure 21. However,the engine specific impulse assumed is more repre-sentative of a Mach 6 ramjet. Comparison withan SST fleet is made in table 4 with the assumptionof 300 billion revenue passengermiles annuallyand an average passenger load factor of 0.5.

The flux of NOx is small, but the concern isover the possibility of a self-regenerativereac-tion discussed previously. The lack of CO2 fluxis a significant benefit of the HST over the SST..However, the flux of water vapor is at most ofthe same order of magnitude as the natural flux

9

transferredfrom the troposphere. The estimateof Manabe and Wetherald2° would indicateapossible increase in the Earth's surface tempera-ture of no more than 1.0°F (0.6°C).

SST HST

Cruise Lift-Drag Ratio 7.75 5.0

Engine Specific Impulse-sec 2200 3840

Average Cruise Wt - lb 500,000 800,000

Cruise Range n.mi. 4000 4000

Payload - Passengers 275 400

Cruise Mach Number 2.7 6.0

FuelFlow Rate - lb/sec 29.4 41.6

TotalFuelPer Flight - lb 273,000 169,000

Revenue Passenger-Miles 300x109 300x109

Load Factor 0.5 0.5

lb NOx/Passenger-Mile .62x10-2 .485x10-2

lb CO2/Passenger-Mile .78 0

lb H20/Passenger-Mile .32 .86

NOx Flux - lb/sec 118 92

CO2 Flux -lb/sec 14,900 0

H20 Flux - lb/sec 6100 16,350

TABLE 4 POLLUTION CHARACTERISTICS-SST AND HST FLEETS

Sonic BoomThe sonic boom problem for an HST occurs pri-

marily during the climb and accelerationportionof the flight. The relativelyhigh cross-sectionalarea of the all-body type of vehicle (figure 1)leads to sonic booms as high as 5 psf (24 kg/m2)transonically,7 whereas the slenderwing-body air-craft will have a sonic boom of about 3 psf (15kg/m2) transonically. Flying higher trajectories,to minimize the transonic boom, leads to oversizedengines and uneconomicalsystems. The use ofrockets to boost to higher trajectorieswas investi-gated for the all-body aircraft and does not appearattractive.7

The situation at cruise, on the other hand, isvery favorable. The high altitudecruise (approxi-mately 108,000 ft (33 km) at Mach 6) results incruise sonic booms of less than 1.0 psf (5 kg/m2)which may be acceptable for overland flight. Manylogical overland routes involvecities locatednear the oceans, and this leads to the possibilityof climbing and descending over water with overlandcruise legs at hypersonic speeds.9 For citieslocated far from the oceans, the only alternativeis subsonic cruise over land followedor precededby hypersonic cruise over the oceans. Figures 22and 23 show the economic penaltiesassociatedwiththese operational alternatives. The penalties forturns during climb and/or descent include theadditionaltime and fuel consumedduring the turn,compared to great circle trajectories,and theadditional time and fuel needed to make up therange normally attained during the climb and de-scent. A trajectory optimizationprogram was usedto minimize fuel consumptionduring the turningascent and descent while maintaininga maximum nor-mal load factor of 1.2. Figure 22 shows the DOCand ROI penalties related to these turns. Alsoshown is the reduction in attainablerange for agiven takeoff gross weight due to the increase infuel consumed.

Figure 22. Effect of Turns to Avoid OverlarldSonic Boom on HST Economics

The economic penalty is significant: for aturn at takeoff the ROI is decreased by about 4%and for turns at both takeoffand descent theROI decreases by about 6%. On the other hand,an increase in passengerload factor to 60% wouldmore than compensatefor the turn penalties,and such a load factor is not unlikely if, forexample, travel time between Los Angeles andNew York is reduced from 5 hours to 2 hours.

Likewise, as shown in figure 23, the eco-nomic penalties for subsoniccruise at thebeginning or ending of the flight are signifi-cant. Cruising 500 n. mi. (930 km) at the endof the flight decreasesthe ROI by approximately5%. For 1000-n. mi. (1850-km)subsonic range theROI decreases by 8%. There is also a slight de-crease in the maximum range for a given takeoffgross weight. If the subsonic leg is at the begin-ning of the flight, both range and economic penal-ties increase due to the higher aircraft weightduring this portion of the flight. Again, theeconomics are marginal,but increases in passengerload factor might compensate. Clearly, all flightsmust have relatively long hypersonic cruise legsto provide a truly attractivereduction in triptime and draw a good load factor.

In summary, the environmentalproblemsconnected with an HST seem potentiallylesssevere than those of an SST aircraft. Theacceleratorengines of an HST are large, toprovide thrust to overcome transonicdrag, andthus may be throttledto give relatively low

1.5

•••,

YE'1.0

o0

a,

L.;.5

E 2.0

76ci 40

30_I1+20m.4 u, 1 I

2 4 6 0 2 4 6

RANGE,B.m,x103, 1 1 I I I I I 1

0 4 8 12 0 4 8 12

RANGE, km x I03

TAKEOFFGROSSWEIGHT,lb o 1,000.000 0 916,000 A 863,000

ksa

a,

a, 60

80

- NOTURNPENALTIES

- - TURNPENALTYATTAKEOFF

- TURNPENALTIESFORBOTHTAKEOFFANDDESCENT

Aso, SUBSONICRANGE

TAKEOFFCROSSeElOIT,lb 0 1,000,000 0 916,000 A 863,000

oRsue

4

3.15

in 2

ci 0 d

E 2.0

.6 1.S

1.0

6

0

-RSUB

801000 \

1000 •

500 •

60

0

•

4

- SISSONICCRUISEAT HID OFFLIGHT

- - SISSONICCRUISEATBEGINNINGOFFLIGHT

0 2 4 6 0 2

RANGE, n. m .x1031 I I , 1 1 1 I I I 1

0 4 8 12 0 4 8 12

RANGE, km x I03

Figure 23. Effect of Subsonic Cruise Rangeon HST Economics

OF0

1,-20

5

5001000

1000

10

noise levels at the airport. Exhaust pollutionfrom an HST is a minor problem in the vicinityof the airport due to the absence of carbonin the fuel and the low cycle pressureof theengine which means low NOx emissions.

In the stratosphere,the potentialproblems are fairly well defined, but themagnitudesof these problems are open toquestion. An HST emits no carbon dioxide, incontrast to the large amounts that will beproduced by a JP-fueled SST fleet, but the amountof water vapor produced by an HST fleet is almostthree times that for an SST fleet. The HST watervapor production is of the same order of magnitudethat occurs naturally, and the effect on the earth'sclimate may be significant. The effects of NOxemissions are the biggest questionmark to date.An HST fleet or an SST fleet will emit comparableamounts of NOx, and the magnitudeseers smallcompared to thewatervaporor carbon dioxideemissions. However, amounts that occur naturallyare unknown, and there is the possibilityof a self-regenerativereaction between nitric oxide and ozonewhich would magnify the problem.

Sonic boom is a problem for any aircraftflying above the speed of sound, and currentlythere appears to be no complete solution to theproblem. However, an HST will create a significantboom only during the climb and acceleration;at cruise the sonic boom overpressurewill beless than 1.0 psf (5 kg/m2) which may be acceptablefor overland cruise.

Developmentof a productionHST is perhapstwo decades away and considerablebasic researchremains to be accomplished.It is much too earlyto decide the fate of future hypersonic transportsbased on environmentalconsiderations. Work isprogressingon the environmentalproblem, andwith the introductionof operationalSST aircraft,these questions should be answered long beforethe initiationof a hypersonic transportdevelop-ment program.

References

Weber, R. J.: "Propulsionfor HypersonicTransport Aircraft," 4th Congress Interna-tional Council AeronauticalSciences,Aug. 1964.

Gregory, T. J.; Petersen, R. H.; andWyss, J. A.: "PerformanceTradeoffs andResearch Problems for Hypersonic Transports,"AIAA Journal of Aircraft, Vol. 2, No. 4,July-Aug. 1965, pp. 266-271.

Jarlett, F. E.: "PerformancePotential ofHydrogen-Fueled,AirbreathingCruise Air-craft," Reports GDC-DCB-66-004/1/2/2A/3/4,General Dynamics/Convair,Sept. 1966, NASAContract NAS2-3180.

Heald, E. R.: "HypersonicTransport DesignConsiderations,"Douglas Paper 3973, present-ed at the Annual Meeting of Japan's Societyof Aeronautical and Space Sciences, Tokyo,Japan, April 6-7, 1966.

Eggers, A. J., Jr.; Petersen, R. H.; andCohen, N. B.: "HypersonicAircraft Tech-nology and Applications,"Astronautics &Aeronautics, Vol. 8, No. 6, June 1970,pp. 30-41.

Hunter, M., II and Fellenz, D. W.: "TheHypersonic Transport--TheTechnology and thePotential,"AIAA Preprint No. 70-1218,Oct. 1970.

Gregory, T. J.; Ardema, M. D.; andWaters, M. H.: "HypersonicTransportPreliminary PerformanceEstimates foran All-Body Configuration,"AIAA Paper70-1224, Oct. 1970.

Becker, J. V.: "Prospectsfor Actively CooledHypersonic Transports,"Astronautics andAeronautics, Vol. 9, No. 8, Aug. 1971,pp. 32-39.

Miller, R. H.: "ThinkingHypersonic," Astronautics and Aeronautics, Vol. 9,No. 8, Aug. 1971, pp. 40-44.

Becker, J. V. and Kirkham, F. S.: "HypersonicTransports," Proceedingsof the NASA Confer-ence on Vehicle Technology for Civil Avia-tion, NASA SP-292, Nov. 1971.

Ardema, M. D.: "StructuralWeight Analysisof HypersonicAircraft," NASA TN D-6692,March 1972.

Waters, M. H.: "Turbojet-RamjetPropulsionSystem for All-Body HypersonicAircraft,"NASA TN D-5993, Jan. 1971.

Wilcox, D. E.; Smith, C. L.; Totten, J. C.;and Hallett, N. C.: "Future Cost of LiquidHydrogen for Use As an Aircraft Fuel," Avia-tion and Space - Progress and Prospects,American Society of Mechanical Engineers,June 1968, pp. 471-478.

Williams, L. O.: "The Cleaning of America,"Astronautics and Aeronautics, Vol. 10, No. 2,Feb. 1972, pp. 42-51.

Lipfert, F. W.: "Correlationof Gas TurbineEmission Data," ASME Paper No. 72-GT-60,March 1972.

Fletcher, R. S.; Siegel, R. D.; andBustress, E. K.: "The Control of Oxides ofNitrogen Emissionsfrom Aircraft Gas TurbineEngines - Volume I: Program Description andResults," U. S. Departmentof TransportationReport No. FAA-RD-71-111,1,Dec. 1971.

Frisken, W. R.: "Extended Industrial Revolu-tion and Climate Change," Transactionsofthe American GeophysicalUnion, Vol. 52,No. 7, July 1971, pp. 500-508.

Anon.: "AtmosphericEffects of SupersonicAircraft," AustralianAcademy of ScienceReport No. 15, Feb. 1972.

Newell, R. E.: "The Global Circulation ofAtmospheric Pollutants,"Scientific American,Vol. 244, No. 1, Jan. 1971, pp. 32-42.

11

Manabe, S. and Wetherald,R. T.: "ThermalEquilibriumof the Atmospherewith a GivenDistributionof RelativeHumidity,"Journalof the AtmosphericSciences,Vol. 24, No. 3,May 1967, pp. 241-259.

Swihart, J. M.: "The United States SST andAir Quality," SAE Paper 710320, Feb. 1971.

Ashby, R. W.; Shimazaki,T.; andWeinman, J. A.: "Effectof Water Vaporand Oxides of Nitrogen on the Compositionof the Stratosphere,"presentedat theInternationalConferenceon Aerospace andAeronauticalMeteorology,Washington,D. C.,May 22-26, 1972.

Newell, R. E.: "Water Vapor Pollution in theStratosphereby the SupersonicTransporter,"Nature, Vol. 226, April 4, 1970, pp. 70-71.

Weickmann, H. K. and Van Vallin, C. C.:"The Sources and Sinks of Water Vapor in theUpper Atmosphere,"presentedat the Inter-national Conferenceon Aerospace andAeronauticalMeteorology,Washington, D. C.May 22-26, 1972.

Johnston, H.: "Reductionof StratosphericOzone by Nitrogen Oxide Catalysts from Super-sonic Transport Exhaust,"Science, Vol. 173,Aug. 1971, pp. 517-522.

Crutzen, P. J.: "Ozone ProductionRates in anOxygen-Hydrogen-NitrogenOxide Atmosphere,"Journal of GeophysicalResearch,Vol. 76,Oct. 20, 1971, pp. 7311-7327.

Henry, J. R. and Beach, H. L.: "HypersonicAir-BreathingPropulsionSystems," Pro-ceedings of the NASA Conferenceon VehicleTechnology for Civil Aviation,NASA SP-292,Nov. 1971.