Upload
dohanh
View
221
Download
2
Embed Size (px)
Citation preview
Proceedings of the 4 `h International Brazing and Soldering Conference, April 26-29. 2009. Hilton in the Walt Disney World Resort, Orlando, Florida. USA
Development of High Temperature Dissimilar Joint Technologyfor Fission Surface Power Systems
Ivan E. Locci''"', Cheryl L. Bowman', and Timothy P. Gabbs'Structures and Materials Division, NASA Glenn Research Center,
21000 Brookpark Rd, MS 49-1, Cleveland, OH 44135University of Toledo, Toledo, Ohio
Ivan.E.L000iLaXASA.GOT"
Abstract
NASA is developing fission surface power (FSP)system technology as a potential option for use onthe surface of the moon or Mars. The goal is todesign a robust system that takes full advantage ofexisting materials data bases. One of the keycomponents of the power conversion system is thehot-side Heat Exchanger (HX). One possibledesign for this heat exchanger requires a joint of thedissimilar metals 316L stainless steel and Inconel718, which must sustain extended operation at hightemperatures. This study compares two joiningtechniques, brazing and diffusion bonding, in thecontext of forming the requisite stainless steel tosuperalloy joint. The microstructures produced bybrazing and diffusion bonding, the effect of brazingcycle on the mechanical tensile properties of thealloys, and the strength of several brazed joints willbe discussed.
Introduction
The U S. Space Exploration Policy calls for lunarexploration activities to enable sustained human androbotic exploration of Mars and more distantdestinations in the solar system. One major concernfor every space exploration mission is the amount ofpower required to sustain life and efficiently operatethe communication and scientific equipment.NASA is evaluating a wide range of powertechnologies that can satisfy anticipated futurespace mission power requirements. Futuregenerations of power systems for spacecraft and
lunar surface systems may require a diverse suite ofpower options that goes beyond the currentcapabilities of chemical, radioisotope thermoelectricgenerator and solar technologies. Recently, NASAhas embarked on a Fission Surface Power (FSP)technology project to develop an option that can beused to provide power for human outposts. The FSPsystem would use nuclear energy converted intoelectricity through a conversion system similar toterrestrial power plants. Key to the success of thisprogram will be the ability to design a compact androbust system based on existing technologies [I].
This is a system with many diverse materials needs.Stainless steel is the preferred structural material inthe reactor section and nickel based superalloys aredesired in the power conversion section. Oneproposed way to reconcile these requirements is tohave a dissimilar metals joint in the hot-side HeatExchanger (HX). For this design, a liquid metalNaK mixture would be flowing on the outside of the316 L stainless steel manifold while transferringheat to the Inconel 718 convertor heater headcontaining pressurized helium gas. The anticipatedjoint temperature is 830 to 870 K (1034 to 1106 °F).Several brazing filler metals applicable to eitherstainless steels or superalloys have been reported[2], but no open literature reports have been founddescribing the integration of both alloy systems.The integrity of the dissimilar metal joint is crucialsince it is in the heat transfer path. In this paper, wediscuss two joining options for this transition fromstainless steel 316 L (SS316L) to Inconel 718(IN718) to fabricate the HX. Both candidateprocesses, diffusion bonding and brazing, do not
16.5
https://ntrs.nasa.gov/search.jsp?R=20090042362 2018-05-31T01:44:37+00:00Z
involve any melting of the base metals and couldprovide an acceptable fabrication bonding processfor the HX application.
Experimental Procedure
The first joining approach used a diffusion bondingtechnique using a hot pressing process. This processwas recently applied [3] to identify diffusionalinteraction issues between various refractory alloyswith superalloys, and SS316L with commercialpurity titanium. The results of the studysuccessfully aided in the process to downselectsuperalloys and refractory alloy candidates.Diffusion bonding was demonstrated to be a goodprimary screening technique to evaluate thediffusional stability of joined dissimilar metals.Also, the process allows the introduction of metalinterlayers of different compositions to limit theinterdiffusion of elements between alloys.
The second joining approach utilized a brazingprocess, where two types of brazing materials wereheated to temperatures approximately 323 K (122°F) above the melting point of the filler material.
The chemistry of the alloys used in this study islisted in Table 1. IN718 is a precipitation-hardenable nickel-chromium alloy containingsignificant amounts of iron, niobium, andmolybdenum along with lesser amounts ofaluminum and titanium. It combines corrosionresistance and high strength with outstandingweldability including resistance to postweldcracking. The alloy has excellent creep-rupturestrength at temperatures up to 973 K (1292 °F).IN718 derives its high temperature strength fromthe precipitation of specific secondary phases (e.g.gamma prime and gamma double-prime) into the
metal matrix_ The precipitation of these nickel-(aluminum, titanium, and niobium) phases isinduced by an age heat treatment in the temperaturerange of 866 - 1088 K (1099 - 1499 °F). For thismetallurgical reaction to properly take place, theaging constituents (aluminum, titanium, andniobium) must be fully in solution.
SS316L is part of the family of austenitic stainlesssteels which constitute the largest stainless family interms of number of alloys and usage. The Ldesignation indicates a lower carbon specification.The main difference between stainless steel type316 from the basic type 304 is the addition ofmolybdenum up to a maximum of 3 wt. %, whichenhances its corrosive properties.
Hot Press Diffusion Bonding Approach
Joints between the superalloy IN718 and theSS316L were produced by hot pressing smallcoupons of each material in vacuum for 2 hours at1000 K (1340 °F) or 1150 K (1610 °F) with apressure of 90 MPa (13 ksi). The 1150 K (1610 °F)temperature was initially selected to ensure thatsome solid state bonding would occur, while the1000 K (1340 °F) temperature was explored tominimize the formation of undesirable secondphases. All samples were surface ground to 600grit, ultrasonic cleaned in acetone and methanol andair dried. Figure 1 shows an illustration of adissimilar material stack-up. By stacking the disks,a single hot-press run could form several diffusioncouples at once. Thin sheets of Ni and V wereexplored as possible metal interlayers between theSS316L and the IN718 alloys. Metal interlayersmay enhanced the bonding and minimize alloy toalloy diffusion reactions.
Table 1. Chemical Analysis of Alloys Studied
Wt.°'o Al B C Co Cr Cu Fe NIn MO Nb Ni Si Ti V
IN 718 0.6 100ppm 0.04 0.12 18.7 540ppm 19.2 870ppm 2.74 4.49 52.9 470ppm 0.89 660ppm
SS316L 50ppm 0.02 0.08 17.1 0.42 68.1 1.42 1.86 80ppm 10.12 0.66 0.024 0.045
166
Couple FabricationDiffusion couple "sandwiches"
stacked together
IN718 V foil 1_^SS316 — Ni foil
IN718
—SS316
1^FIGURE 1. Illustration of Hot Pressing Sequenceto Produce Multiple, Dissimilar Metal DiffusionCouples.
Brazing Approach
The brazing approach is another joining process thatdoes not involve any melting of the base metal. Thecoalescence of the metal surfaces to be joined isgenerally produced by the distribution of a lowermelting point filler metal by capillary attraction tothe closely fitted surfaces, by heating to suitabletemperatures [4]. Properly designed braze joints canremain structurally intact and hermetically soundunder heavy pressures, even when the joint issubjected to shock or vibrational types of loading[5]. The compatibility of several filler materialswith the SS316L or IN718 were initially evaluatedusing an overlapping type joint technique. Figure 2shows schematic representations of the overlappingbrazing process for flat surfaces using either foil or
SS316— Brazing SS316 —
Alloy
in Brazing
IN718 — Paste Forth
IN Foil
BrazingAlloy in
IN718 Paste Form
SS316—
Before Brazing ^ After Brazing
FIGURE 2. Illustration of various BrazingApproaches using Paste and Foil to Evaluate theProcess and the Compatibility of variousConunercial Braze Alloys with SS316L andIN71 S alloys.
paste, and the capillarity attraction process fortightly fitted SS316L pipe and IN718 solid rod.Based on the projected use temperature, 830 - 870K (1034 - 1106 °F), two classes of brazing materialswere initially selected, nickel- and gold-basedalloys. The nickel-based braze alloys, listed inTable 2, have similar characteristics and elementsalready present in the bonding alloys. The gold-based alloys listed may be less reactive, but aremore expensive. Both brazing systems have similarcoefficient of thermal expansion as the substratealloys. Other critical aspects such as melting point,the ability to vet the surfaces of the materials toform a continuous, sound and strong bond seem tobe adequate.
Table 2 — Brazing Materials Evaluated and Post-Bonding Thermal Exposures
Nickel-basedBrazes
Composition (pit. %)Brazing Temperature Post-Bonding
Exposures at 900 K
AMDRY 790 N1-1.7413-3.22S1 1403 K (2066 °F) 100, 1000, 3000 h
AMDRY 108 Ni-23Cr-11.5Fe-4.2P-6.4Si 1403 K (2066 °F) 100, 1000, 3000 h
AMDRY 775 Ni-15.38Cr-3.813 1403 K (2066 °F) 100, 1000, 3000 h
Gold-based Brazes
Nioro 82Au-18Ni1258 K (1805 °F)1273 K (1832 °F)
100, 1000, 3000 h
Palniro-7 70Au-22Ni-8Pd 1323 K (1922 °F) 100, 1000, 3000 h
167
Method for Evaluating Alloyand Brazed Joints Strength
The evaluation of the strength of brazed jointsfollowed the recommended practice from theAmerican Welding Society (AWS), AmericanNational Standard C3.2M/C3.2:2008 [6]. Since thetwo alloys evaluated have different strength, whereIN718 is typically much stronger than the SS316L,two sheets of SS316L were used to double-upagainst a single sheet of IN718 to fabricate multipledouble-lap shear specimens. All the faying surfaceswere prepared with a 600 grit finish, ultrasonicallycleaned with acetone and alcohol. The tensiontesting of the double-lap shear specimens wereperformed in accordance with ASTM E 8, StandardTest Methods for Tension Testing of MetallicMaterials. All specimens were tested in tension at830 K (1034 °F) by Metcut Research Inc.(Cincinnati, OH). The breaking load and locationof the failure, whether through the brazed area or inthe metal base, were recorded.
The average shear stress in the filler metal at failureand the average tensile stress in the base metal atfailure were computed using the equation from theAWS standard as shown:
Average shear stress (brazing filler metal)Breaking load / (2A x W) (1)
and,
Average tensile stress (base metal)Breaking Load / (W x T) (2)
Where, A is the brazed joint overlap length, W isthe width, and T is the material thickness (all inmm). Control specimens without brazed joints,processed through the brazing and secondary heattreatment cycles, were also tested at 830 K (1034°F) to determine the effect of the thermal cycles onthe properties of the metals. Figure 3 shows typicalbrazed double-lap shear and control specimens used
FIGURE 3. Typical Brazed Double-Lap Shear andControl Specimens Used for Testing at 830K.
in this study. Base metal hardness measurementswere performed on all the samples to monitor theeffect and reproducibility of the brazing heattreatment process.
Results and Discussion
Diffusion Bonded Joint
Observations of the bond produced under varioushot pressing conditions were used to guide theselection of the threshold temperature to create asound bond with minimal reaction zone. Initialbonding experiments indicated that hot pressing at1150 K (1610 °F) for 2 hours produced a soundbond line. Table 3 lists the diffusion couples thatwere formed by diffusion bonding as well assubsequent annealing conditions. Figure 4 showsoptical micrographs of as-bonded couples (a)SS316L to IN718, (b) SS316L to 25 µm thick Ni-foil interlayer to IN718, and (c) SS316L to 25 µmthick V-foil interlayer to IN718. Bonding occurredin all three systems; however it is quite evident thatthe V-foil has reacted and embrittled the interfacebetween the two alloys and resulted in nearlycontinuous longitudinal cracks. (Fig. 4c).
168
-interlayer
Table 3. Diffusion Bonded Couples EvaluatedHot Press Bonding Conditions and Post-Bonding
Thermal Exposures
Hot PressingConditions
ExposureTemp, K
ExposureTime, h
1150 K (877 °C)for 2 h at 90 MPa
SS316L/IN718 1000 300, 1000
SS316L/Ni/IN718 1000 300, 1000
SS316LN/IN718 1000 300, 1000
1000 K (727 °C)for 7 h at 90 MPa
SS316L/IN718 1000 j 300.1000
The Ni-interlayer appears to accommodategracefully the bonding of both alloys, althoughsome fine porosity as shown in the Field EmissionSEM (FESEM) image (Fig. 5), were observed at theNi/IN718 interface. Energy dispersive spectrumanalyses confirmed the presence of Al 203 particlestrapped in the porosity which are believed to becoming from the grinding media used duringmetallographic sample preparation. Similar porositywith the presence of small Al203 particles was alsodetected at the interface formed in the hot pressedSS316L to IN718 couple. A hot isostatic pressing(HIP) process which can provide up to three timesthe pressure compared to hot pressing, is beinginvestigated, to understand if the porosity is due tothe limited hot pressing pressure, or if it was formedby the preferential diffusion of some elementalspecies present in the alloys which can leave behindKirkendall porosity.
Figure 4. Optical Micrographs of Hot Pressed at1150 K for 2 hours at 90 MPa Couples: (a) SS316Lto IN718, (b) SS316 to Ni- interlayer to IN316, and(c) SS316L to V-interlayer to IN718. BondInterface Indicated with Arrows.
169
Figure 6. FESEM BSE Images of Nioro BrazedSS316L to IN718 in the (a) as-Brazed Condition,and after Long Term Exposures at 900 K after (b)100, (c) 1000 h, (d) 3000 Hours.
170
IN 718
x2D.Ok GWBSE 61291200700
Figure 5. FESEM Image and Energy DispersiveSpectrum Confirming the Presence of Al2O3Particles Trapped in the Fine Pores Formed at theIN718/Ni-Interlayer Interface after Hot Pressingat 1150 K for 2 Hours at 90 MPa.
Br aZed Joints
.Vicr•ostr•uctur•a 1 Observations
As discussed previously, two classes of commercialbrazing materials were evaluated. The chemicalcompatibility and stability of the braze alloys andthe two different base metals were observed afterbrazing and long term exposures. Figure 6 showsFESEM backscattered images of Nioro brazedSS316L to IN718 after brazing for 5 minutes at1248 K (1787 °F) in vacuum at 1.33 x 10 -3 Pa, andafter post-brazing exposures at 900 K (1160 °F) for100, 1000 and 3000 hours. Minimal reaction with
either alloys and the filler material, or change in thebrazed microstructure was observed.
In general, a good bond was accomplished with orwithout Ni-plating. Figure 7 shows a micrograph ofnickel-plated brazed SS316L to IN718 with Nioroafter a 3000 h exposure at 900 K (1160 °F). Ni-enriched areas were observed at either side of thebraze joint. Occasionally small regions withdiscontinuous porosity were detected in themetallographic cross sections. Figure 8 presentselemental maps which show the distribution of thevarious elements present in the IN718 and SS316Lalloys and in the brazed zone after brazing withNioro, and a short anneal heat treatment at 900 K(1160 °F) in argon for 1000 hours. Diffusionprocesses can change the composition of the brazejoint and also the chemical and physical propertiesof the boundary interface between the joined alloysand the filler metal. From the maps, it is clear thatgold has not diffused into either alloy. Some Ni/Auenriched strings associated with grain boundariesare observed at the SS316/Braze interface, although
10 wt.% of Ni is already present in the SS316base composition. Some indication of Fe presenceis observed in the center of the braze, at the samelocation where Ni-enriched pockets are visible.Both alloys have comparable amounts of Cr, but noindication of Cr diffusion into the brazed joint wasdetected.
Figure 9 shows that the Ni-based brazing fillermetals listed in Table 2 have the ability of to wetand spread to both SS316L and IN718 alloys,characteristics required to form a continuous, soundand strong bond. FESEM observations and EDSanalyses of the various phases forined with the Ni-brazes have revealed some degree of reaction nearthe interface and along the base metal grainboundaries. Brazing with AMDRY 775 (AM775)resulted in the apparent segregation of Cr at grainboundaries near the SS316L / braze interface, whileNb and Mo enrichment occurred at the grainboundaries near the IN718 / braze interface.
Figure 7. FESEM BSE Image of Ni-platedSS316L and IN718 plates brazed with Nioroafter a 3000 Hours Exposure at 900 K.
Figure 8. FESEM BSE image of Nioro brazedSS316L to IN718 after long terin exposure at900 K for 1000 hours with energy dispersivespectroscopy elemental maps for Au, Ni, Fe, andCr.
Chromium segregated in the center of the braze aspart of a eutectic formation.Brazing with AMDRY 790 (AM790) revealed asimilar enrichment of Cr at SS316L / braze interfaceand Nb at IN718 / braze interface. Well defined
171
55]16 - 1._{
`^Y
indication of Cr-enrichment was detected in theSS316 side; however possible enrichment of Nb orMo occurred at the interface. Further clarification,if needed, would require other techniques, since theP-Ka, Nb-La, and Mo-La x-ray peaks overlap.
Mechanical Properties
IN :1s
Figure 9. FESEM Images Showing ApproximateLocation of the Original Interfaces, Features of theBrazing Bonds and Reaction near the BondInterfaces after the Brazing Process using (a)AM775, (b) AM790, and (c) AM108 Brazes.
grains, with grain boundaries enriched in Ni anddepleted in Si, can be clearly visualized inside thebrazed section. Brazing with AMDRY 108(AM108) resulted in a complex microstructure dueto the presence of the higher numbers of elementsinitially available in the AM108 braze. No clear
The mechanical performance of a brazed joint isdetermined by the integration of the braze fillermetal, the base metal, the geometry of thebrazement, and the brazing procedure [6]. First theresponse of the base metal to the perspective braze-cycle thernial exposures was investigated since thebase metals properties, especially for the IN718, arevery sensitive to the thermal cycle processparameters. Tensile testing close to the targettemperature of 830 K (1034 °F) resulted in asignificant drop in the ultimate and yield strength,and considerable increase in the elongation forIN718 after the simulated brazing exposures. Nosignificant changes were observed in the ultimatestrength or elongation for the SS316, but a slightdrop (-10%) in the yield strength was noticed.Figure 10 summarizes the effects of various thermalcycles. Again, the goal was to understand theramifications of the brazing thermal cycle on themechanical properties. Therefore sample strengthsare compared in the starting condition, after abrazing-like heat treatment without the presence ofbraze alloy, after a brazing-like heat treatment withthe presence of braze alloy, as well as the braze-likeheat treatment followed by an aging step. Theappropriate starting condition was solution anneal +aging for the IN718 and as-received for theSS316L. The representative brazing cycle was aNioro-type brazing at 1248 K (1787 °F). Thestrengths of the stainless steel were largelyunchanged but it appeared that the superalloy wasadversely affected by the representative brazethermal cycle. It was found that the tensile strengthof the solution anneal + aged IN718 can berecovered with a secondary aging heat treatmentafter a Nioro-type brazing cycle.
172
1200IN718 q UTS
a1000
— n 0.2% YS2 600
0iY 600g SS316L—'n 400
200
0
aCF, Io Aoa
¢ x P °0 `y^\ Ax PA ec, ¢xPG O rG1^c, `9ti ^ Goa
oAG ae
°\^
PcccA
O-a^
^
0a^
°A ^AO^aT i%5
5P% ^Px
FIGURE 10. Effect of the Brazing, Anneal, AgeThermal Cycles, and Contact with Braze Materialon the Mechanical Properties of IN718 andSS316L at 830K.
After detennining appropriate brazed and post-brazed thermal treatments, the tensile strength of thebrazed joints were evaluated with double-lap shearspecimens. As recommended by the standardmethod, various overlaps were explored as well asvarious types of brazes. Figure 11 shows typicalcross sections of double lap specimens with 1.5 and9 mm overlaps. In most cases the brazing materialcompletely filled the gap between the plates. Figure12a shows a plot of the average shear stress,calculated with equation 1, and the average tensilestress in the base metal, calculated with equation 2,at 830 K with Nioro as a brazing filler metal, asfunction of overlap distance. Typically as theoverlap is increased, the failure shifts from failurein the braze joint to failure in the base metal. Inother words, joints made with an overlap greaterthan this transition value will behave as beingstronger than the materials being joined. In the caseof the Nioro braze, the overlaps studied were notsufficient to reach this transition value. Failuresoccurred in the filler material independently if thematerial was or not exposed to a post-aging heattreatment. However, as shown in Figure 12b, wherethe average shear stress and tensile stress obtainedfor a Palniro-7 braze are plotted, the 9 mm overlapresulted in the shift of failure from the brazed jointto the SS316 base metal; the filled symbolsrepresent failure through the base metal.
Figure 11. Cross Sections of SS316L/IN718 Double-lap Shear Specimens Brazedwith Nioro with (a) 1.5 and (b) 9 mmOverlaps.
900Only failure m the Nioro filler a
900a
700
600
500q ©--- Shear - As-Brazedm 400 Shear - Brazed + AgedN 300 fs Tensile - As-Brazed
200—^ Tensile - Brazed + Aged
mc 100
0
0 1 2 3 4 56 7 a 9 10Overlap Distance, mm
900 b
a 800
0 700
= 600C Shear-As-Brazed
Shear - Brazed + Agedm 500
4 Tensile - As-Brazed0 400 ^ Tensile - Brazed +Agedn 300A 200o Open symbols = failure in the filler materiala 100
Filled Symbols = failure in the SS 316L Alloy0
0 1 2 3 4 6 6 7 a 9 10Overlap Distance, mm
Figure 12. Average Shear Stress and TensileStress as a Function of Overlap Distance with (a)Nioro: Failure Occurred only in the Brazing FillerMetal, and (b) Palniro-7: Failure Transition fromFiller to SS316L Base Metal at the 9 mm Overlap.
173
Nioro Palniro-7 Ni-based,& 1000a0
m 800
Q
m 600
U)m 400
U)
rn 200
Q 0
Figure 13 contrasts the two modes of failuresobserved with the same 9 mm overlap, throughthe SS316L base metal brazed with Palniro 7 (Fig.13a), and through the Nioro filler material (Fig.13b). The average tensile stress at failure and theaverage shear stress at failure are plotted in Figure14 as a function of braze materials and overlapdistances. The double-lap specimens brazed withPalniro 7 show consistently higher tensile stressvalues than the Nioro or the Ni-based brazes.
9 lnm Overlap with Palniro-7
SS316L
IN718
Din a
9 mm Overlap with Nioro
SS316L
f' IN718
SS316L
b
Figure 13. Different Types of Failures with same9 mm Overlap (a) Failure in the Base Metal withPalniro-7 (SS316L), and (b) Failure in the NioroBrazing Filler Metal.
Conclusions
In order to ensure a metallurgically sound joint thatwould survive extended use at approximately 830 K(1034 °F), diffusion bonding and brazing of SS316Land IN718 were studied. The joining of twodissimilar metals required an evaluation of the
1000 <
CD
mv
800 CCD
Cn600 (D
CDcn400 Sn
v200 =
v0 n
t° ^ oo oo
Q^ ^ Q^. Q^
alb g^^o2 2 Qts P
Figure 14. Average Shear and Tensile Stress atFailure as a Function of Braze Materials andOverlap Distances. All Brazed Specimens WereAged before Testing.
effect of the pressing parameters or the brazingcycles and post joining heat treatments on themicrostructures and mechanical properties of bothalloys. Hot press diffusion bonding trials showedthat bonding occurred with and without a purenickel interlayer. Fine porosity was detected at thenew interfaces, but may be reduced or eliminatedwith higher pressures and optimized bondingparameters. Micro structural observations indicatedthat gold-based brazes resulted in minimal reactionwith either base metals. The chemical reaction wasobserved to be more extensive with any of the N1_based brazes studied. Double-lap shear tensilespecimens were successfally fabricated and tested at830 K (1034 °F). For a specific overlap, theaverage tensile strength observed from the brazeddouble-lap shear specimens for Palniro 7 and theNi-base brazes are higher compared to the Niorobraze; however, the higher brazing temperaturesrequired for Palniro 7 and any of the Ni-brazes mayimpact negatively on the properties of both basemetals. The Inconel 718, which is extensively usedin the aerospace industry, required only an agingheat treatment, after a typical Nioro brazing cycle,to regain most of its strength. The stainless steel316L alloy, not affected from secondary phaseprecipitation or dissolution, maintained its ultimate
174
tensile strength after a typical Nioro brazing cycle,however its yield strength was more sensitive to thethermal exposures. Based on the lowest impact onthe base metal strength and the minimal chemicalinteraction, Nioro is the leading braze candidate tojoin the two dissimilar metals for the HXapplication.
Acknowledgments
This research was supported through the FissionSurface Power (FSP) technology project under theNASA Exploration Systems Mission Directorate.The author gratefully acknowledges the excellenttechnical assistance of Dr. James Nesbitt, DonaldHumphrey in the hot pressing processing lab, JoyBuehler in the metallographic lab, Terry McCue inthe microscopy lab, Marc Jaster, John Juhas andAdrienne Veverka in the furnace room, summerstudent Sara Caruso, and Timothy Ubienski, JosephLavelle, Aldo Panzanella, Anthony Kapucinski inthe specimen machining group.
References
[1] Mason, L., Poston, D., Quall, L., "SystemConcepts for affordable Fission Surface Power",NASA/TM-2008-215166.
[2] Lucas, M. J. and Manente, D., ASM HandbookVol. 6: Welding, Braziny, and Soldering, ASMInternational, Materials Park, OH, 2005, pp.912-930.
[3] Locci, I. E., Nesbitt, J. A., Ritzert, F. J., andBowman, C. L., "High Temperature Stability ofDissimilar Metal Joints in Fission SurfacePower Systems", in Space Technology, andApplications International Forum-STAIF 2007,El-Genk, M. S., ed., 2007, pp 260-667.
[4] Schwartz, M. M., Brazing, ASM International,Materials Park, OH, 2nd edition, 2004, pp. 7-13.
[5] Schwartz, M. M., reference 2, pp. 114-125.[6] Standard Method for Evaluating the Strength of
Brazed Joints AWS C3.2M/C3.2:2008, 01edition, American Welding Society, Miami,Florida.
17.5