Citation
Refractory Carbides' Microstructural Integrity in Hot Hydrogen Environment of Space Nuclear Reactors

Material Information

Title:
Refractory Carbides' Microstructural Integrity in Hot Hydrogen Environment of Space Nuclear Reactors
Creator:
Cunningham, Brandon
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (69 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Nuclear Engineering Sciences
Nuclear and Radiological Engineering
Committee Chair:
Anghaie, Samim
Committee Members:
Dugan, Edward T.
Dempere, Luisa A.
Graduation Date:
8/9/2008

Subjects

Subjects / Keywords:
Carbides ( jstor )
Carbon ( jstor )
Corrosion ( jstor )
Hereditary hemorrhagic telangiectasia ( jstor )
Hydrogen ( jstor )
Images ( jstor )
Music analysis ( jstor )
Solid solutions ( jstor )
Specimens ( jstor )
Uranium ( jstor )
Nuclear and Radiological Engineering -- Dissertations, Academic -- UF
carbide, epma, fuel, hydrogen, microstructure, nuclear, propellant, propulsion, reaction, sem, space, temperature, xrd
Genre:
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Nuclear Engineering Sciences thesis, M.S.

Notes

Abstract:
This project required the testing of three mono-carbides to investigate which carbide has the least degradation of microstructural properties when exposed to a hot hydrogen environment. This was performed in order to determine the usability of various carbides in a base mixture. The purpose of this research was to investigate carbides as candidate fuel matrix for nuclear thermal propulsion. Tantalum carbide (TaC), tungsten carbide (WC), and zirconium carbide (ZrC) were all tested in a hot hydrogen environment at an average temperature of 2775K. Each study tracked the carbon content, corrosion, density, hydrogen embrittlement, and phase changes. The results show that some carbides experienced changes in all of the aforementioned properties, while others experienced a combination of changes or no noticeable degradation. The second stage of the research tested solid solution uranium bearing tri-carbides in a hot hydrogen environment. The tri-carbides included the compositions (U0.1, Zr0.58, Nb0.32)C0.95, (U0.1, Zr0.68, Nb0.22)C0.95, (U0.1, Zr0.77, Nb0.13)C0.95, and (U0.05, Zr0.62, Nb0.33)C0.95. All tri-carbides were exposed to temperatures above 2900K for two hours. Scanning electron microscopy, bulk carbon analysis, X-ray diffraction, density measurements, and mass measurements were used to characterize each specimen. Analysis showed that differences in metal composition had no noticeable effect on final fuel integrity. Carbon analysis, electron microscopy, and diffraction analysis showed that tri-carbides displayed little to no change in microstructure due to exposure to a hot hydrogen environment. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2008.
Local:
Adviser: Anghaie, Samim.
Statement of Responsibility:
by Brandon Cunningham.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Cunningham, Brandon. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:

cunningham_b.pdf

cunningham_b_Page_19.txt

cunningham_b_Page_36.txt

cunningham_b_Page_22.txt

cunningham_b_Page_28.txt

cunningham_b_Page_06.txt

cunningham_b_Page_32.txt

cunningham_b_Page_17.txt

cunningham_b_Page_24.txt

cunningham_b_Page_49.txt

cunningham_b_Page_20.txt

cunningham_b_Page_56.txt

cunningham_b_Page_23.txt

cunningham_b_Page_44.txt

cunningham_b_Page_16.txt

cunningham_b_Page_66.txt

cunningham_b_Page_11.txt

cunningham_b_Page_13.txt

cunningham_b_pdf.txt

cunningham_b_Page_29.txt

cunningham_b_Page_46.txt

cunningham_b_Page_27.txt

cunningham_b_Page_51.txt

cunningham_b_Page_18.txt

cunningham_b_Page_58.txt

cunningham_b_Page_07.txt

cunningham_b_Page_48.txt

cunningham_b_Page_61.txt

cunningham_b_Page_03.txt

cunningham_b_Page_34.txt

cunningham_b_Page_64.txt

cunningham_b_Page_45.txt

cunningham_b_Page_04.txt

cunningham_b_Page_42.txt

cunningham_b_Page_40.txt

cunningham_b_Page_33.txt

cunningham_b_Page_31.txt

cunningham_b_Page_62.txt

cunningham_b_Page_25.txt

cunningham_b_Page_55.txt

cunningham_b_Page_30.txt

cunningham_b_Page_53.txt

cunningham_b_Page_43.txt

cunningham_b_Page_65.txt

cunningham_b_Page_26.txt

cunningham_b_Page_37.txt

cunningham_b_Page_02.txt

cunningham_b_Page_50.txt

cunningham_b_Page_67.txt

cunningham_b_Page_15.txt

cunningham_b_Page_38.txt

cunningham_b_Page_08.txt

cunningham_b_Page_60.txt

cunningham_b_Page_47.txt

cunningham_b_Page_41.txt

cunningham_b_Page_14.txt

cunningham_b_Page_21.txt

cunningham_b_Page_63.txt

cunningham_b_Page_52.txt

cunningham_b_Page_35.txt

cunningham_b_Page_09.txt

cunningham_b_Page_68.txt

cunningham_b_Page_54.txt

cunningham_b_Page_12.txt

cunningham_b_Page_59.txt

cunningham_b_Page_05.txt

cunningham_b_Page_10.txt

cunningham_b_Page_01.txt

cunningham_b_Page_69.txt

cunningham_b_Page_57.txt

cunningham_b_Page_39.txt


Full Text





displayed constant carbon content from pre- to post-HHT status. The fourth tri-carbide sample

reported a loss in carbon content. This was most likely due to carbon contamination during

processing and handling of the specimen. The SEM images showed no regions of

microstructural segregation with particular attention given to the surface since the surface is

where the depletion of carbon would be the greatest. The XRD patterns verified the SEM and

bulk carbon analysis by showing that the tri-carbide specimens remained as unaffected solid

solution carbides. Based on the analysis, solid solution uranium tri-carbides appear to be

compatible with hydrogen to a maximum tested temperature of 2926K for two hours.

Hydrogen corrosion due to temperature is only one mechanism that can affect the rate at

which hydrogen can attack a fuel matrix. Future work investigating carbides and more

importantly tri-carbides as NTP fuel should examine the effects of other corrosion mechanism

such as fuel irradiation. Since the hot hydrogen corrosion process consists of temperature effects

coupled with radiation exposure and thermal mechanical stresses, future tests should attempt to

incorporate multiple corrosion mechanisms.






































Figure 3-3. Copper cooled induction coil with tungsten susceptor assembly











-tC Norm w%




7.75 C-wt% Pre-HHT


620 645
Distance (microns)


695 720


Figure 4-17. Mid-axial cross section of post-HHT showing WC transition zone of carbon content




-Pre-HHT -Post-HHT


1

0.9

0.8

0.7

' 0.6
C
rc
S0.5

0.4
o

0.3

0.2

0.1

0


25 30 35 40 45 50 55 60 65
29 (deg.)


Figure 4-18. WC pre- and post-HHT XRD spectrum


545 570


JI

. . . . . .











Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

REFRACTORY CARBIDES' MICROSTRUCTURAL INTEGRITY IN HOT HYDROGEN
ENVIRONMENT OF SPACE NUCLEAR REACTORS
By

Brandon Warren Cunningham

August 2008

Chair: Samim Anghaie
Major: Nuclear and Radiological Engineering

This project required the testing of three mono-carbides to investigate which carbide has

the least degradation of microstructural properties when exposed to a hot hydrogen environment.

This was performed in order to determine the usability of various carbides in a base mixture.

The purpose of this research was to investigate carbides as candidate fuel matrix for nuclear

thermal propulsion. Tantalum carbide (TaC), tungsten carbide (WC), and zirconium carbide

(ZrC) were all tested in a hot hydrogen environment at an average temperature of 2775K. Each

study tracked the carbon content, corrosion, density, hydrogen embrittlement, and phase changes.

The results show that some carbides experienced changes in all of the aforementioned properties,

while others experienced a combination of changes or no noticeable degradation.

The second stage of the research tested solid solution uranium bearing tri-carbides in a hot

hydrogen environment. The tri-carbides included the compositions (Uo.1, Zro.58, Nb0.32)C0.95,

(Uo.1, Zro.68, Nbo.22)Co.95, (Uo.1, Zro.77, Nb0.13)C0.95, and (Uo.os, Zr0.62, Nb0.33)C0.95. All tri-carbides

were exposed to temperatures above 2900K for two hours. Scanning electron microscopy, bulk

carbon analysis, X-ray diffraction, density measurements, and mass measurements were used to

characterize each specimen. Analysis showed that differences in metal composition had no









LIST OF REFERENCES


[1]. T. W. Knight, "Processing of Solid Solution Mixed Uranium/Refractor Metal Carbides
for Advanced Space Nuclear Power and Propulsion Systems" Ph.D. Dissertation,
University of Florida, Gainesville, FL 2000.

[2]. S. Anghaie, T. W. Knight, "Development of Robust Tri-carbide Fueled Reactors for
Multi-Megawatt Space Power and Propulsion Applications", American Nuclear Society
2004 Winter Meeting, Washington D.C., 2004.

[3]. D. P. Butt, D. G. Pelaccio, M. S. El-Genk, "A Review of Carbide Fuel Corrosion for
Nuclear Thermal Propulsion Applications", 11th Symposium of Space Nuclear Power and
Propulsion Conference, Albuquerque, NM, 1994.

[4]. Pewee Reactor Test Report, Los Alamos National Laboratory August 1969.

[5]. D. R. Koenig, "Experience Gained from the Space Nuclear Rocket Program (Rover)",
Los Alamos National Laboratory, May, 1986.

[6]. D. G. Pelaccio, M. S. El-Genk, D. P. Butt, "Hydrogen Corrosion Consideration of
Carbide Fuel for Nuclear Thermal Propulsion Applications", Journal of Propulsion and
Power, Vol. 11 No6, November 1995.

[7]. S. I. Sandlers, Chemical and Engineering Thermodynamics 3rd Ed., John Wiley & Sons,
Inc., NY, 1999.

[8]. NIST-JANAF Thermochemical Tables, 3RD Ed., 1985.

[9]. NIST-JANAF Thermochemical Tables, 4TH Ed., 1998.

[10]. W. F. Seng, P. A. Barnes, "Calculations of Tungsten Silicide and Carbide Formation on
SiC using the Gibbs Free Energy" Material Science and Engineering B, Elsevier, Vol. 72
Issue 1, March 2000.

[11]. JCPDS International Centre for X-Ray Diffraction Data. StandardX-Ray Diffraction
Powder Patterns from the JCPDS Research Associateship. s.l., Newtown Square, PA,
1996.

[12]. Boeing Aerospace Company, "Advanced Propulsion Systems Concepts for Orbital
Transfer Study", NASA MSFC, Huntsville, AL,1981

[13]. T. W. Knight, S. Anghaie, "Processing and Fabrication of Mixed Uranium/Refractory
Metal Carbide Fuels with Liquid-Phase Sintering", Journal of Nuclear Materials,
Elsevier, Vol. 306, 2002

[14]. B. Panda, R. R. Hickman, S. Sandeep, Solid Solution Carbides are the Key Fuels for
Future Nuclear Thermal Propulsion", NASA MSFC, Huntsville, AL, 2005
































Figure 4-42. TRI-C2 BSE image of pre-HHT specimen


Figure 4-43. TRI-C2 BSE image of post-HHT specimen


























S9 LO I
3 i





I -4



0LS O.9 LO lIl
CARBON ATOM RATIO, x*C/I+2r

Figure 2-2. Pseudo-binary phase diagram of (Uo.i,Zro.9)Cx (Lyon, 1973)































Figure 4-3. TaC BSE image showing post-HHT cracking


Figure 4-4. TaC SE image showing fracture zone












-Pre-HHT -Post-HHT


6O00 -




5000



4000 -


-. 0.


-0.5 0 0.5 1 1.5 2 2.5
Energy (keV)


Figure 4-1. EDS of pre- and post-HHT TaC sample


3 3.5 4 4.5 5


Figure 4-2. Optical image of TaC. A) Pre-HHT TaC sample. B) Post-HHT TaC sample


Ta





















Ta
01


3000




2000




1000


~Lc~ -- ---,-



#1












17.400



17.200



17.000



16.800



c 16.600


10400rr


Post-HHT, 16.775






Pre-HHT, 16.450


16.200



16.000



15.800


Figure 4-25. ZrC bulk carbon content.


20.00

18.00


16.00

14.00

12.00

10.00


8.00

6.00

4.00

2.00

n nn


-_ I


--post -E-pre
-i


0 50 100 150 200 250 300 350 400
stance (mirons)


Figure 4-26. Mid-axial cross section showing ZrC carbon content as a function of depth.


\NV m _,__


-

-

-


-

-

-


































2008 Brandon Warren Cunningham































Figure 4-23. ZrC BSE image of pre-HHT specimen


Figure 4-24. ZrC BSE image of post-HHT specimen



















Figure 4-11. Optical image of WC. A) Pre-HHT WC sample. B) Post-HHT WC sample


Figure 4-12. WC SE image of pre-HHT specimen












ri~


J .
-tir -rt'tnI






C. : -.. .
w `rz-





PT 4..













Figure 4-15. WC BSE image showing the transition zone where carbon depletion occurred




-4-post -I!- pre
-ftY






I


~, ..;.IZ: .r,..




ij1 r Vt~:




Figure 4 15. WC BSE image showing the transition zone where carbon depletion occurred




-4-post -C-pme


20.00

18.00

16.00

14.00

12.00

10.00

8.00

6.00


17 ---~A


0 50 100


150 200 250 300 350 400 450


Distance (microns)


Figure 4-16. Mid-axial cross section showing WC carbon content as a function of depth









Table 3-1. Average testing conditions for mono-carbide samples
TaC WC ZrC
Mass Flow Rate 15.58 15.19 15.74 8/

Temperature 2762 2779 2776 Kelvin

Chamber Pressure 968 969 970 torr

Run Time 2 2 1.167* hr

*The ZrC sample test was conducted for a shorter time because of equipment constraints.


Table 3-2. Average testing conditions for tri-carbide samples
(Uo.l,Zro.58,Nbo.32)Co.95 (Uo.1,Zro.68,Nbo.22)Co.95 (Uo.1,Zro.77,Nbo. 13)C.95 (Uo.05,Zro.62,Nbo.33)Co.95
Mass Flow Rate 15.81 14.92 14.92 14.92 g/

Temperature 2926 2903 2903 2903 kelvin

Chamber Pressure 982 964 964 964 torr

Run Time 2 2 2 2 hr































Figure 4-5. TaC SE image of pre-HHT specimen


Figure 4-6. TaC SE image of post-HHT specimen









CHAPTER 2
BACKGROUND AND LITERATURE REVIEW

2.1 Historical Relevance for Fuel Development

Nuclear reactors for rocket applications were first ground tested during the Rover program.

The main purpose of the Rover program was to develop nuclear powered rockets for ballistic

missile defense. When the need for nuclear rocket applications diminished due to the efficiency

of chemical rockets, Los Alamos National Laboratory (LANL) contracted with Westinghouse

Electric Corporation to create a research and development program within the original

framework of the Rover program. The new program, the Nuclear Engine for Rocket Vehicle

Applications (NERVA), was developed for space purposes. The NERVA program was charged

not only with reactor design, but also to achieve the highest possible propellant temperature. The

intent was a completed rocket design with maximized ISP while maintaining low weight.

Equation 1 defines ISP as proportional to the square root of the fluid temperature over the

fluid molecular weight [2]. Hydrogen is the de facto coolant for NTP because its molecular

weight is the lowest and highest specific heat of all elements. All parameters being equal, the

temperature of the coolant dictates the efficiency of the rocket design. Therefore, fuel

development was rigorously pursued in order to meet this goal. Throughout the course of the

Rover program several reactors series with varying fuel forms were investigated. Figure 2-1

illustrates the fuels used during the Rover/NERVA program. Nuclear reactor cores developed

during these programs generally operated with temperatures around 2500K and pressures of

3MPa. The series that is most important for this research, the Pewee and Nuclear Furnace

families, focused on alternative fuels, higher fuel temperature, and fuel performance.









Hydrogen


CVD Coating -
SUC2 Particle

PyC Coating -

LC2 Sphere -.4


a) UC2 Particles Dispersed
in Graphite Mrtrix


GiOphite
"0 Sutbstrate 0
0 -0 -
o 0
--+




b) PyC Coated UC2 Spheres
Dispersed in Graphite


ZrC Coating


Giaphite
Matrix

, (U,Zr)C ..


Hydrogen


c) Composite Fuel d) Solid Solution (U,Zr)C

Figure 2-1. Four fuel forms used in Rover/NERVA designs (Matthews et al., 1991)


Grapflte- /
(--Se&esfrate
C> >
o O ^

c m o 0'i<& C''2


q


Hydrogen


~cc~-ci-










20LIm


Figure 4-21. ZrC SE image of pre-HHT specimen


Figure 4-22. ZrC SE image of post-HHT specimen
























IA
.1 4~


Figure 4-44. TRI-C3 BSE image of pre-HHT specimen


Figure 4-45. TRI-C3 SE image of post-HHT specimen


































To the spark within humanity to defy conventional wisdom.









ternary carbide system stability was made by observing the Gibbs potential for mono-carbides

since uranium tri-carbide is a solid solution of (U,Zr,X)C where X is a metal.

The second law of thermodynamics was used to determine whether the mechanism for

microstructural damage to mono-carbides was predominantly corrosion or erosion. The Standard

Free Energy of Formation (AGf) gives the work released or consumed by a chemical reaction at

constant temperature and pressure. If AGf is positive the chemical reaction is non-spontaneous

and if AGf is negative the chemical reaction is spontaneous. A positive AGf means that the

species reaction cannot take place under the given conditions without the minimum of the AGf

calculated [7]. The AGf was calculated for several chemical reactions that are possible in a

hydrogen environment, neglecting the presence of trace elements such as oxygen in the testing

system. The dissociation energy of diatomic hydrogen is 4.5eV. This is higher than the thermal

energy of the system, 2700K or 0.232eV. For simplification, only atomic hydrogen would be

considered because of its chemical volatility. Figure 2-3 displays the Gibbs free energy plot for

several reactions. With the exception of one chemical reaction each of the other reactions are not

possible at the temperatures of interest without the addition of more work. This implies that the

main mechanism for microstructural damage is erosion rather than corrosion. The one chemical

reaction that is spontaneous is WC+4H--W+CH4, which readily occurs at temperatures beneath

445K or its equilibrium temperature. This is a reaction that is plausible for corrosion to occur

and carbon loss to be significant. Using this principle the mono-carbide, WC, should experience

a noticeable loss in carbon content. Since erosion is the mechanism for damaging the TaC and

ZrC matrix those compounds could exhibit a uniform loss of both metal and carbon.












-Pre-HHT -Post-HHT


-0.19 1.01 2.21




Figure 4-48. TRI-C1 pre- and post-HHT EDS


3.41 4.61


-Pre-HHT -Post-HHT


u

C'


-0.19 1.01 2.21

Energy (KeV)


Figure 4-49. TRI-C2 pre-and post-HHT EDS.


3.41 4.61


























Figure 4-30. Optical image of TRI-C3. A) TRI-C3 pre-HHT. B) TRI-C3 post-HHT


Figure 4-31. Optical image of TRI-C4. A) TRI-C4 pre-HHT. B) TRI-C4 post-HHT









4-17 Mid-axial cross section of post-HHT showing WC transition zone of carbon content......47

4-18 W C pre- and post-HHT XRD spectrum ........................................ ......................... 47

4-19 ED S of pre- and post-HHT of ZrC sam ple ......... ..................... .... .....................48

4-20 Optical image of ZrC. A) Pre-HHT ZrC sample. B) Post-HHT ZrC sample ...................48

4-21 ZrC SE im age of pre-H H T specim en .............................................. ....... .... ...............49

4-22 ZrC SE im age of post-HHT specim en........................................ ............................ 49

4-23 ZrC BSE image of pre-HHT specimen................................ .... ........... 50

4-24 ZrC BSE image of post-HHT specimen ......................... ......... ............... 50

4-25 Z rC bulk carbon content. ........................................................................... ....................51

4-26 Mid-axial cross section showing ZrC carbon content as a function of depth.................51

4-27 ZrC pre- and post-HHT XRD spectrum ........................................ ......................... 52

4-28 Optical image of TRI-C1. A) TRI-C1 pre-HHT. B) TRI-C1 post-HHT .........................53

4-29 Optical image of TRI-C2. A) TRI-C2 pre-HHT. B) TRI-C2 post-HHT .........................53

4-30 Optical image of TRI-C3. A) TRI-C3 pre-HHT. B) TRI-C3 post-HHT .........................54

4-31 Optical image of TRI-C4. A) TRI-C4 pre-HHT. B) TRI-C4 post-HHT .........................54

4-32 TRI-C1 pre- and post-HHT bulk carbon content .............. .............. ..................55

4-33 TRI-C2 pre- and post-HHT bulk carbon content ...................................................55

4-34 TRI-C3 pre- and post-HHT bulk carbon content ...................................................56

4-35 TRI-C4 pre- and post-HHT bulk carbon content............... .............. .....................56

4-36 T R I-C 1 X R D spectrum ..............................................................................................57

4-37 TR I-C 2 X R D spectrum ............................................................................... ............... 57

4-38 T R I-C 3 X R D spectrum ..............................................................................................58

4-39 TR I-C 4 X R D spectrum ............................................................................... ............... 58

4-40 TRI-C1 BSE image of pre-HHT specimen................................. ................59

4-41 TRI-C1 BSE image of post-HHT specimen ........... .......................... 59












8.400


8.300


8.200


8.100


8.000


7.900


7.800


SPre-HHT, 8.136


7.700
7.6 TaC bulk carbon content


Figure 4-7. TaC bulk carbon content


-4-post --pre
20.00

18.00

16.00
U
FT
c 14.00

S12.00
U
u
a 10.00
N



6.00

4.00

2.00

0 .0 0 .ii
0 50 100 150 200 250 300 350 400
Distance (microns)


Figure 4-8. Mid-axial cross section showing TaC carbon content as a function of depth


I Post-HHT, 7.975


I









ACKNOWLEDGMENTS

I first want to thank my mother, Tafferlon Ann Cunningham; and my grandparents, Minnie

Lee and Willie Bud White, Sr. They have been and continue to be instrumental in making me

the person I am today. I thank the supervisory committee members, Dr. Edward Dugan and Dr.

Luisa Amelia Dempere. Special gratitude goes to my supervisory committee chair, Dr. Samim

Anghaie, for giving me the opportunity to conduct this research. Dr Anghaie's knowledge of

nuclear engineering is noteworthy of a true genius.

All experiments were conducted at the Applied Ultra High Temperature Research

Laboratory of the Innovative Nuclear Space Power and Propulsion Institute (INSPI) at the

University of Florida. Analysis was performed at the Major Analytical Instrumentation Center at

the University of Florida.






























Figure 3-1. Hot hydrogen test chamber


Figure 3-2. Top view of sample/tungsten susceptor/inductive coil configuration









CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS

Testing the mono-carbides, WC, TaC, and ZrC, in a hot hydrogen environment yielded

different results depending on the carbide composition. The sample that had the least change in

microstructural integrity was the ZrC sample. The sample's carbon content remained the same.

No cracking, grain growth, density changes, or hydrogen embrittlement was observed. The XRD

spectrum showed the post-HHT specimen to have no change in phase from the pre-HHT

specimen.

The TaC post-HHT sample performed the second best in having minimum degradation on

the microstructural integrity. There was no depletion of carbon from the sample, both bulk and

cross-sectional; however, the sample did experience extensive cracking. A density decrease of

4.836% in the post-HHT sample was measured against the pre-HHT sample.

The WC sample performed the worst in a hot hydrogen environment. The post-HHT

sample experienced a noticeable degree of phase change, grain boundary separation, carbon

depletion, and possible formation hydrocarbons. Portions of the post-HHT sample's XRD

spectrum showed the formation of W2C. W2C most likely formed on the edge where carbon

depletion occurred. After a two hour test at 2775K, the hydrogen was shown to deplete carbon

from the WC matrix at rate of 5.25 Pmin WC also experienced the greatest change in density,

decreasing by 10.91%. Based on these results WC would not be a safe fuel candidate for NTP

because of the threat of fuel failure due to changes in the microstructure.

Testing of the four solid solution uranium tri-carbides was also conducted at higher

temperatures (approximately 2900K) for two hours, representative of NTP operating time.

Analysis showed the tri-carbides displayed little to no evidence of changes due to the hot

hydrogen environment. The bulk carbon analysis showed that three of the four carbides









The immersion density test showed that for the TaC pre-HHT sample the density was

14.350 /M3 The post-HHT sample had 13.656 g/ 3 yielding a percent change of 4.836%.


The reference theoretical density for TaC is 14.498 g 3 making the pre-HHT sample 98.98%

dense and the post-HHT sample 94.19% dense after a two hour run time. Since pores are often

the sites for crack initiation it is possible that pores are the cause of the extensive cracking

throughout the TaC sample.

The XRD analysis for the pre- and post-HHT showed that the structure did not undergo a

phase change. Figure 4-9 displays the results. If any of the reactions in figure 2-3 would have

occurred, peaks indicating elemental tantalum would have been present.

4.1.2 Tungsten Carbide

The EDS displayed in figure 4-10 indicates that the elemental compositions for the pre-

and post-HHT samples are consistent. The other analysis for the post-HHT sample shows a large

difference in properties from the pre-HHT. The blistering on the surface in figure 4-1 lb

indicates the formation of a liquid or a gas. Assuming there were trace amounts of oxygen in the

system and the formation of tungsten oxide was negligible, the blistering on the surface could be

attributed to trapped methane gas. The post-HHT WC sample shows that the degradation

occurred at the grain boundaries. An explanation for the blistering can attributed to the Gibbs

potential calculated for WC. Hydrogen reacted with carbon in the WC matrix forming methane

or hydrocarbons. That methane gas then diffuses though the path of least resistance or the

boundaries to the surface. The gas that did not escape was trapped under the surface to blister.

Significant grain growth occurred in the WC sample. Figure 4-12 and figure 4-13 show the pre-

HHT and post-HHT grain structure respectively. This can be attributed to conducting the testing

near the melting point of tungsten carbide (3143K).









The XRD spectrum shows that there were significant changes in phase. The pre-HHT

sample in figure 14-18 illustrates that only the WC phase is present. The post-HHT sample

reveals the presence of two phases: WC and W2C. The presence of the W2C is evidence that the

carbon-to-metal ratio has decreased due to depletion of carbon from the microstructure surface.

4.1.3 Zirconium Carbide

The EDS for the ZrC pre- and post-HHT samples are displayed figure 4-19. The results

show that both samples' composition is consistent throughout the experiment. Figure 4-20a and

4-20b shows that on the surface the post-HHT sample displays little to no changes when

compared to the pre-HHT sample. There is a slight discoloration caused by hot hydrogen

etching the carbonaceous or loose material on the surface. Figures 4-21 and 4-22 show that

qualitatively the microstructure was unchanged. The BSE images in 4-23 and 4-24 show no

compositional differences or any depleted carbon regions due to hydrogen flow.

A bulk carbon analysis confirmed the qualitative result of an unchanged post-HHT ZrC

microstructure. Figure 4-25 illustrates the pre-HHT C-wt% valued at 16.450wt% and the post-

HHT C-wt% valued at 16.775wtO%. Though the post-HHT sample is higher, the values are

within close statistical error. The theoretical ZrC carbon content is 11.634wt%. The EMPA in

figure 4-26 shows that there was no depletion of carbon at the surface of the sample. The

variability in data points is caused by surface roughness.

Changes in density could reveal the presence of hydrogen embrittlement. The pre-HHT

sample's density was 6.648 /C 3, while the post-HHT sample's density was 6.635 gc/ The

change in density from the control sample is 0.197%. Given the precision of the immersion

testing, the values of the pre- and post-HHT density are statistically the same.









LIST OF TABLES

Table page

3-1 Average testing conditions for mono-carbide samples.................................................28

3-2 Average testing conditions for tri-carbide samples ................................. ............... 28

4-1 U uranium tri-carbide results ....................................................................... ................... 38

4-2 Uranium tri-carbide pre- and post-HHT carbon content.................................................38









L IST O F R E FE R EN C E S ....................................................... ...................................67

B IO G R A PH IC A L SK E T C H .............................................................................. .....................69




















































6












(Uo., Zro.s, Nbo.s2)Co.s
-Pre-HHT -Post-HHT


25 30 35 40 45 50 55 50 655
20 (deg.)


Figure 4-36. TRI-C1 XRD spectrum





(Uo.1, Zro.e- Nbo.22)Co.
-Pre-HHT -Post-HHT


30 35


40 45 50
20 (deg.)


55 60 65


Figure 4-37. TRI-C2 XRD spectrum


I


. ... J 'Lis









noticeable effect on final fuel integrity. Carbon analysis, electron microscopy, and diffraction

analysis showed that tri-carbides displayed little to no change in microstructure due to exposure

to a hot hydrogen environment.









LIST OF ABBREVIATIONS

BSE Backscattered Electron

C-wt% Carbon Weight percent

EDS Energy Dispersive Spectrum

EMPA Electron Micro Probe Analysis

HHT Hot Hydrogen Testing

HIP Hot Isostatic Press

INSPI Innovative Nuclear Space Power and Propulsion Institute

ISP Specific Impulse

LANL Los Alamos National Laboratory

MWth Megawatt thermal

NASA MSFC National Aeronautics and Space Administration Marshall Space Flight Center

NEP Nuclear Electric Propulsion

NERVA Nuclear Engine Rocket Vehicle Application

NF Nuclear Furnace

NTP Nuclear Thermal Propulsion

PyC Pyrolitic-Carbon

SE Secondary Electron

SEM Scanning Electron Microscopy

TaC Tantalum Carbide

WC Tungsten Carbide

XRD X-Ray Diffraction

ZrC Zirconium Carbide

AGf Standard free energy of formation









Reactor concept or Neptune reactor concept, it acts as a coolant. Uranium tri-carbide was chosen

for testing because preliminary results showed that tri-carbides have higher melting temperatures

and better resistance to hot hydrogen corrosion than the binary-carbide fuel (U,Zr)C [2].

Testing materials in a high temperature hydrogen environment is rare because most

operating conditions do not warrant its use. As a result, background information and literature is

limited. The main concern with NTP is hot hydrogen corrosion. Hydrogen at high temperatures

can erode carbon from the matrix or react and corrode the fuel substrate to form methane or other

hydrocarbons. Studies of previous NTP programs and case studies have shown that the hot

hydrogen corrosion mechanisms are a highly coupled and highly complex process. Research

shows that carbide's material properties are dependent on maintaining stoichiometry [3].

Changes in the fuel's carbon-to-metal ratio can lead to changes in the liquidus, changes in phase

orientation, fatigue resistance, and toughness. A quantification of carbon-to-metal ratio with

exposure time to temperature is needed to ensure properties such as melting point, thermal

conductivity, and consistent thermal expansion are coherent throughout the fuel's anticipated

lifetime. Evidence of hot hydrogen corrosion can show as hydrogen-induced blistering, hot

embrittlement, mass loss, and hydride formation.









corrosion cracking, which consequently creates a larger surface area and deeper route for the hot

hydrogen to attack. The corrosion-induced cracks and pores weaken the microstructure and

allow for surface erosion. The depletion of carbon from the fuel surface leads to changes in local

composition, which alters local material properties. In solid solution uranium bi-carbides it has

been shown that material properties depend heavily on the carbon-to-metal ratio. Figure 2-2

shows a phase diagram of a pseudo-binary carbide (Uo.1, Zro.9)Cx. In this composition the

carbon-to-metal ratio is limited to 0.92-0.96 before a solid fuel transforms to a mixture of

liquid/solid solution carbide. Allowing this to occur will ultimately lead to fuel failure.

Fuel irradiation enhances the rate of hot hydrogen corrosion. Exposure of the

microstructure to radiation, particularly fast neutrons, creates lattice dislocations and point

defects. The localized defect effectively reduces the local thermal conductivity which produces a

thermal gradient with the surrounding lattice sites. The defects also reduce local ductility. This,

combined with the reduction in thermal conductivity, produces thermal stress-induced cracking

that exposes more surface area to hot hydrogen attack.

Creep is a mechanism that does not enhance the hot hydrogen corrosion rate, unlike the

other two processes. The operating temperatures of NTP fuel accelerate the creep process.

Creep will serve to anneal the localized high stress zones deterring the formation of crack growth

and closing micro-cracks within the microstructure.

2.2.2 Free Energy of Formation Calculations

Binary carbides and ternary carbides are materials whose full use has yet to be

investigated. As a result, there is no tabulated data on the free energy of formation to determine

the system's thermodynamic stability. The free energy of formation for the mono-carbides was

calculated to determine the system's stability in a hydrogen environment. An approximation of









CHAPTER 4
RESULTS AND DISCUSSION

4.1 Mono-Carbides


4.1.1 Tantalum Carbide

To begin the analysis an energy dispersive spectrum (EDS) was taken to verify that the

sample contains the given elements. Figure 4-1 shows the EDS both pre- and post-HHT. The

post-HHT spectrum was taken to verify there was no contamination-atomic numbers higher

than carbon-were introduced to the sample after HHT. Figure 4-2 gives a view of the pre- and

post-HHT TaC samples. Compared to the pre-HHT sample, the post-HHT shows extensive

cracking on the surface. A cross section was taken through the post sample and subsurface

cracking was also present. Figures 4-3 and 4-4 show the extent of cracking through a mid-axial

cross-section. Figures 4-5 and 4-6 show that the grain structure was unaffected by the presence

of hydrogen. Neither grain growth nor grain degradation occurred, but the grain area adjacent to

the cracks was fractured. Fractures could be attributed to fast heat up rate.

TaC has a theoretical carbon weight percent (C-wt%) of 6.224wt%.. Figure 4-7 shows the

pre-HHT C-wt% at 8.136wt% and post-HHT at 7.975wt%, both of which are hyper-

stoichiometric. Statistically there was no change in carbon content. To determine if that small

difference in the mean C-wt% for the pre- and post-HHT samples is due to the depletion of

carbon at the surface from hydrogen flow, an EMPA was performed. The results of the EMPA

shown in figure 4-8 illustrates that the pre- and post-HHT samples show no change in carbon

concentration from the surface of the sample to within 360tm. The data point at 400tm

represents the center of the sample or 3mm from the edge. The fluctuation in the EMPA profile

is caused by surface roughness. The surface roughness scatters the incident electrons in

directions that exaggerate or decrease the counts in the solid angle of the detector.











Standard Free Energy of Formation
1000.000



900.000 -


800.000


700.000 --' ..


600.000


500.000 -,




1+ +a ------------_----


200.000 -_ ----- -- 70

100.000-- -H#
- -





20.000
3HO 600 900 1200 1500 1800 2100 2400 2700 3000 3300 3600 3900
Temperature (K)
-100.000

Figure 2-3. Standard free energy of formation of possible chemical reactions of mono-carbides in a hydrogen environment [8,9,10]









3.4 Analysis Methodology

The following variables were examined for pre- and post-hot hydrogen testing (HHT):

overall carbon concentration, carbon and metal concentration as a function of depth, density, and

phase and crystallographic changes. Several visual observations such as grain growth, inter and

intra-granular corrosion or erosion, spallation, cracking, swelling, and hydrogen embrittlement

were also made. The pre- and post-HHT density was measured by immersion testing. A LECO

WC-200 carbon determinator was used to combust specimens to investigate the bulk carbon

concentration. A carbon profile of each specimen as a function of depth was characterized by

EMPA. The EMPA instrument used was a JEOL SUPERPROBE 733 Electron Probe.

Topographical and compositional contrast and elemental composition were inspected with a

JEOL JSM-6400 SEM. SEM was conducted to make most visual observations. Using XRD,

phase and crystallographic properties were analyzed with Philips APD 3720 X-Ray

Diffractometer and reference spectra database [11]. Specimen weight and dimensions were

measured with the Sartorius R180D analytic balance with an accuracy of 0.00001g and a

Mitutoyo Absolute Digimatic Caliper (CD-6" CSX), respectively.









-Pre-HHT -Post-HHT


I I


35 40 45 50
28 (deg.)


55 60 65


Figure 4-27. ZrC pre- and post-HHT XRD spectrum


N









Figure 4-14 indicates that the bulk carbon content for the pre-HHT sample was 12.7%

while the post-HHT sample had a C-wt% of 5.928. Since pre-and post-HHT carbon content are

outside of their statistical error then depletion of carbon from the microstructure occurred with

certainty. The large difference in the pre- and post-HHT carbon content was due to carbon

depletion at the surface of the WC in the post-HHT sample. Figure 4-15 shows a BSE image of

the WC post-HHT. The brighter edge region indicates there was higher tungsten content at the

edge. The depletion of carbon at the edge was verified using EMPA. Figure 4-16 shows that the

post-HHT sample has less carbon content within the first 360tm from the edge. The data point

at 400tm represents the center or 3mm from the edge of the sample. That data point at the

center of the sample indicates there is a diffusion gradient of carbon in the post-HHT sample.

Figure 4-17 illustrates that at a depth of 630tm, the WC post-HHT sample's depleted carbon

zone returns to the carbon concentration of the pre-HHT sample. This gives a carbon depletion


depth rate approximately 5.25 /min

Notice the difference between the pre-HHT bulk 12.7C-wt% and EMPA's pre-HHT

average trend which shows a lower C-wt%. Figure 4-15 shows dark spherical regions rich in

carbon that would have been included in the bulk carbon analysis but not the EMPA result.

Therefore the C-wt% for the WC matrix follows the EMPA trend, average 7.75wt%, but the

sample's C-wt% is 12.7%. These rich carbon regions were probably introduced during the

sample's fabrication as contamination. There is a large difference in the pre- and post-HHT

density. The pre-HHT density is 15.873 3 while the post-HHT sample is 14.141 3
/ hl Urrl potHTsml s1.11/ c.Tr


This gives a percent difference of 10.91%.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S ................................................................................................ . 7

LIST OF FIGURES .................................. .. ..... ..... ................. .8

L IST O F A B B R E V IA T IO N S .............................................................................. ... ............... 11

A B S T R A C T ......... ....................... ............................................................ 12

CHAPTER

1 IN T R O D U C T IO N ............................... .................. ...................................... .................... 14

2 BACKGROUND AND LITERATURE REVIEW ..................................... ...............16

2.1 Historical Relevance for Fuel Development ............. .....................................16
2 .1.1 P ew ee S series ............................................ ... ................................... ..... 17
2.1.2 N nuclear Furnace Series ........................................... .................. ............... 18
2.2 Literature R review ...................................................... ... ............ .............. 18
2.2.1 Hot Hydrogen Corrosion Mechanisms............................. ...............18
2.2.2 Free Energy of Formation Calculations............... ............................................19

3 EXPERIMENT METHODS AND PROCEDURES................................... ...............24

3.1 M ass Flow D eterm ination... .............................................................................. 24
3.2 Equipm ent and Testing Procedures ........................................ ........................ 24
3.3 Specim en Preparation for A nalysis......................................... ......................... 25
3.4 A analysis M methodology ........... ......... ...... ....................................... .........................27

4 RESULTS AND DISCUSSION............... ....................................... ...............31

4 .1 M on o-C arb ides ................................................................ 3 1
4.1.1 Tantalum Carbide .................. .......................... .... ..................... 31
4 .1.2 T ungsten C arbide............ ... .................................................... .. .... ..... .. 32
4 .1.3 Z irconium C arbide............ ... .................................................. .. .... ..... .. 34
4.2 Tri-Carbides ....................................................................35
4 .2 .1 B ulk C arb on A naly sis............................................................ .....................35
4 .2 .2 X R D A nalysis........... ............................................................ .. .... ..... .. 36
4.2.3 SEM /ED S A analysis ................................................ .............................. 36

5 CONCLUSIONS AND RECOMMENDATIONS ...................................... ............... 65









BIOGRAPHICAL SKETCH

Brandon Warren Cunningham was born in 1983. He was raised in Mount Dora, Florida.

He graduated sum cum laude from Mount Dora High School in May 2001. Brandon attended the

University of Florida and majored in nuclear engineering. During the summer of 2006, Brandon

received a fellowship with the Center of Space Nuclear Research at Idaho National Laboratory.

He graduated with his bachelor's degree from the Nuclear and Radiological Engineering

Department at the University of Florida in December 2006.









REFRACTORY CARBIDES' MICROSTRUCTURAL INTEGRITY IN HOT HYDROGEN
ENVIRONMENT OF SPACE NUCLEAR REACTORS


















By

BRANDON WARREN CUNNINGHAM


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2008















1


0.9


0.8


0.7


0.6


0.5


0.4


0.3


0.2


0.1


0


25 30 35 40 45 50

2B (deg.)


Figure 4-9. TaC pre- and post-HHT XRD spectrum


4000


3500


3000


2500


2000


1500


1000


500


55 60 65


-Pre-HHT -Post-HHT

w


C w


-0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5


Figure 4-10. EDS of pre- and post-HT of WC sample

Figure 4-10. EDS of pre- and post-HHT of WC sample


-Pre-HHT -Post-HHT


















------- ---- ------------------------------
--------------- ------- -------------------------------------------------



-------------- ------- ---------------------------------------------[


I


.









due to carbonaceous material contaminating the surface and filling the pores. Samples TRI-C1,

TRI-C2, and TRI-C3 showed no measurable decrease in carbon content implying little to no

erosion or corrosion of carbon from the tri-carbide matrix occurred. The bulk carbon analysis

demonstrated that the tri-carbides' carbon content remain close to stoichiometric weight percent

4.2.2 XRD Analysis

Another measure of changes in the microstructure would have been revealed through

XRD. None of the post-HHT diffraction patterns in figures 4-36 through 4-39 showed the

presence of different phases from that of the pre-HHT samples' diffraction pattern. The structure

was considered to be a solid solution because the diffraction pattern did not exhibit the presence

of segregated carbide constituents. Observing the XRD pattern ofNbC, ZrC, and UC/UC2

powders from a previous research [1], the diffraction pattern of the tri-carbides verified that a

solid solution of(U,Zr,Nb)C exists in both the pre- and post-HHT specimens. No changes in

phase or crystallographic orientation occurred as a result of the two hour test at 2900K.

4.2.3 SEM/EDS Analysis

Using SEM, compositional contrast and topographical images were obtained. Figures 4-40

through 4-47 are images of the specimens' microstructure. Heterogeneity of Zr and Nb in the

microstructure could not be ascertained from BSE images due to the proximity of their atomic

numbers. It can be inferred that had the pre- and post-HHT XRD patterns displayed ZrC and

NbC spectra, the two elements would have been segregated in the microstructure. Knowing that

this was not the case, Zr and Nb could be assumed to have a homogeneous distribution. Based

on compositional contrast images, all the tri-carbides displayed no regions of depleted carbon or

a gradient of uranium from the surface. In addition to SEM images, EDS was taken of each pre-

and post-HHT sample to ensure no contamination was introduced to the sample during testing or









Table 4-1. Uranium tri-carbide results
Sample cn Pre-HHT Post-HHT Relative Pre-HHT Post-HHT Mass
ame Tri-carbie composition Density Density density Mass Mass Difference ecentage
Designation (atom fraction) ( 3 3 O cf mass loss
(g/cm) (g/cm ) difference (g) (g) (g)
TRI-C1 (U0o. Zro.58,Nb0.32)C0.95 6.1066 5.7567 -5.73% 2.24545 2.11287 -0.13258* -5.9%

TRI-C2 (Uo. 1Zr.68,NbO.22)C095 6.5306 6.1129 -6.40% 2.53249 2.43116 -0.10133* -4.0%

TRI-C3 (Uo.1 Zr,Nb0.13)C0.95 6.2807 6.4215 +2.24% 2.85000 2.81581 -0.03416* -1.2%

TRI-C4 (U0.05, Zr0.62,N0.33)C0.95 6.2961 6.9031 +9.64% 4.83599 4.70004 -0.12561* -2.81%

* A negative value implies a loss occurred. A positive value implies a gain occurred


Table 4-2. Uranium tri-carbide pre- and post-HHT carbon content
Designan Theoretical Carbon Content Pre-HHT Carbon Content Post-HHT Carbon Content
Sample Designation) ( ) ( )
TRI-C1 9.689 10.875 11

TRI-C2 9.700 0.9375 9.9425

TRI-C3 9.710 9.93 9.725

TRI-C4 10.32 11.775 10.35











-Pre-HHT -Post-HHT


1500 -
U






50 c




-0.19 1.01 2.21 3.41 4.61

El.'.l2 T,-V)


Figure 4-50. TRI-C3 pre-and post-HHT EDS.




-Pre-HHT -Post-HHT
3000 z



2500 i



2000


-0.19 1.01 2.21 3.41 4.61

Er,,-gyFigure -C4T E


Figure 4-51. TRI-C4 pre-and post-HHT EDS









CHAPTER 3
EXPERIMENT METHODS AND PROCEDURES

3.1 Mass Flow Determination

A linear relationship was used to ensure that the samples tested in this experiment

experienced a relative mass flow rate similar to a nuclear space reactor. The Pewee reactor from

the Rover/NERVA program was used as a template. The Pewee reactor used a total hydrogen

flow rate of 19.341kg with 14.62kg cooling the fuel. The total available coolant area of

858,929cm2 was based on dimensions of the core [4]. This gives a mass flow per square

centimeter of 7.9401 x 105 k s.c2 The mono-carbide and tri-carbide specimen tested in this

experiment had an available coolant area of approximately 3.5cm2 resulting in an equivalent

mass flow rate of 2.71 x 10 4 k/ The dimensions of each specimen vary slightly making

3.5cm2 the estimated sample surface the coolant contacts. Chamber pressure, temperature, and

mass flow rate were recorded every 10 minutes to calculate and average testing conditions. All

average testing conditions are listed in tables 3-1 and 3-2.

3.2 Equipment and Testing Procedures

Two series of samples were tested. Mono-carbide samples produced at National

Aeronautics and Space Administration's Marshall Space Flight Center (NASA MSFC) and

uranium tri-carbides produced at University of Florida's Innovative Nuclear Space Power and

Propulsion Institute (INSPI). The mono-carbides were fabricated by hot Isostatic press (HIP).

Tri-carbides were fabricated by an induction heated uni-axial hot press liquid phase sintering

method [1]. During this process a homogeneous mixture of three carbides are cold pressed and

heated to a temperature above the lowest melting point carbide constituent, in this case uranium

carbide. The liquid uranium carbide fills in porous areas and contacts the grain boundary from










[15]. D. A. Porter, K. E. Easterling, Phase Transformation in Metals and Alloys 2nd Ed., CRC
Press, Boca Raton, FL, 1992.

[16]. C. R. Wang, J. M.Yang, W. Hoffman, "Thermal Stability of Refractory Carbide/boride
Composites", Materials, Chemistry and Physics, Elsevier, Vol. 72 Issue 3, March 2002.

[17]. R. Zee, B. Chin, J. Cohron, "Hot Hydrogen Testing of Refractory Metals and Ceramics"
Auburn University, Auburn, AL, February 1993.










Isp = AC Eqn. 1
I p= AC -



A = performance factor related to thermo-physical properties of the propellant

Cf= thrust coefficient, which is a function of nozzle parameters

Th = chamber temperature (K)

Me = molecular weight of propellant

F = thrust (N)

m = mass flow rate of propellant (kg/s)


2.1.1 Pewee Series

The Pewee series was intended to be a family of small inexpensive test reactors used for

high temperature fuels demonstration: only one reactor was eventually built and demonstrated.

In 1968 the Pewee 1 ran successfully at 503MWth for 40 minutes with an exit coolant

temperature of 2550K. The peak fuel temperature was measured at 2750K. It demonstrated the


highest power for a reactor of its size, with an average power density of 2340M 3h and a


peak power density of 5200 3 The calculated ISP in vacuum was 845s. Pewee


contained 402 hexagonal fuel elements loaded with pyrolytic-graphite coated UC2 particles

dispersed throughout a graphite matrix. The fuel elements were coated with NbC, ZrC, and Mb

as an overcoat. Inspection of the fuel elements after testing showed that many of the elements

exhibited cracked regions or external corrosion. Fractures were due to thermal-mechanical

stresses, but corrosion was attributed to the fuel elements' graphite matrix being subjected to a

hot hydrogen environment [4].

























Figure 4-28. Optical image of TRI-C1. A) TRI-C1 pre-HHT. B) TRI-C1 post-HHT


Figure 4-29. Optical image of TRI-C2. A) TRI-C2 pre-HHT. B) TRI-C2 post-HHT

































9.6



9.5



9.4


Figure 4-34. TRI-C3 pre- and post-HHT bulk carbon content


Pre-HHT, 11.775


10.8


10.6


10.4
Post-HHT, 10.35

10.2


10


Figure 4-35. TRI-C4 pre- and post-HHT bulk carbon content









to account for the hardness of carbides. Individual sections were taken for X-ray Diffraction

(XRD), bulk carbon analysis, Electron MicroProbe Analysis (EMPA), and Scanning Electron

Microscope (SEM) analysis. All specimens were ultrasonically cleaned and degassed in distilled

water for 60 minutes with a Bransonic B3-R. All specimens were dried in a desiccator for a

minimum of 24 hours before the analysis was conducted.

a. XRD Preparation

i. Specimens were either pulverized to a powder in an alumina crucible or
sectioned to approximately 0.5mm thickness and mounted on glass slides
using a Amyl acetate and Collodian solution.

b. Bulk Carbon Analysis

i. Specimens prepared for bulk carbon analysis were pulverized to a powder
in an alumina crucible.

ii. Three to five batches weighing 0.25g from each sample composition were
placed in a zirconia crucible. High purity accelerator was added to ensure
samples could be inductively heated and ignited. A mixture of 1.8g of
copper and 1.2g of iron were the amounts recommended by the
equipment's operations manual. The iron carbon impurity content was
8PPM.

iii. Samples were placed in carbon analyzer for combustion analysis. The
carbon determinator was calibrated using a tungsten carbide standard with
6.18wt% C.

c. SEM and EMPA Preparation

i. Sections for SEM and EMPA were mounted in phenol ring molds using a
1:5 ratio LECO hardener to LECO epoxy resin, respectively.

ii. LECO Spectrum System 1000 with semi-automatic heads was used for
rough polishing mounted specimens from 180grit to 1200grit.

iii. The final polishing steps decreased surface roughness from 15tm to
0.5[tm. Polishing was performed by hand.

iv. Sections intended for SEM were acid etched for one minute with
HN03/HC1/H2SO4 solution.

v. Sections intended for EMPA was not acid etched.




































Figure 4-13. WC SE image ofpost-HHT specimen
Figure 4-13. WC SE image of post-HHT specimen


14.000


12.000


10.000



8.000
c
..Q

u 6.000


4.000


2.000


0.000

Figure 4-14. WC bulk carbon content


i Pre-HHT, 12.700


SPost-HHT, 5.928












11.1


11.05

Post-HHT, 11.00
11II


10.95


10.9

C Pre-HHT, 10.875
-n
10.85


10.8


10.75


10.7


10.65


Figure 4-32. TRI-C 1 pre- and post-HHT bulk carbon content





10.1




10.05




10




C 9.95
0
g 9.95 -----------------------------------

SIPre-HHT, 9.9375 Post-HHT, 9.9425


9.9




9.85







Figure 4-33. TRI-C2 pre- and post-HHT bulk carbon content










(Uo., Zro.77, Nbo.13)Co.s
-Pre-HHT -Post-HHT


_ 1 1


S 30 35 40 45 50 55 60 65
2e (deg.)


Figure 4-38. TRI-C3 XRD spectrum


(Uo.o, Zro.62, Nbo.33s)C.95
-Pre-HHT -Post-HHT


*11 * +


_I~7


25 30 35 40 45 50
20 (deg.)

Figure 4-39. TRI-C4 XRD spectrum


55 60 65









LIST OF FIGURES


Figure page

2-1 Four fuel forms used in Rover/NERVA designs (Matthews et al., 1991)........................21

2-2 Pseudo-binary phase diagram of (Uo.1,Zro.9)C, (Lyon, 1973) ................. ............ .....22

2-3 Standard free energy of formation of possible chemical reactions of mono-carbides in
a hydrogen environm ent [8,9,10] .......................................................................... .... 23

3-1 H ot hydrogen test cham ber......................................................................... .................. 29

3-2 Top view of sample/tungsten susceptor/inductive coil configuration.............................29

3-3 Copper cooled induction coil with tungsten susceptor assembly ................. ................30

4-1 EDS of pre- and post-HHT TaC sample......... .............. ............ 39

4-2 Optical image of TaC. A) Pre-HHT TaC sample. B) Post-HHT TaC sample .................39

4-3 TaC BSE image showing post-HHT cracking......................................... ............... 40

4-4 TaC SE im age show ing fracture zone ........................................... ......................... 40

4-5 TaC SE image of pre-HHT specimen................. .............................................. 41

4-6 TaC SE im age of post-HHT specim en ........................................ .......................... 41

4-7 T aC bulk carbon content........................................................................... ....................42

4-8 Mid-axial cross section showing TaC carbon content as a function of depth....................42

4-9 TaC pre- and post-HHT XRD spectrum ............................................................ ........... 43

4-10 EDS of pre- and post-HHT of W C sample .............................. .....................43

4-11 Optical image of WC. A) Pre-HHT WC sample. B) Post-HHT WC sample...................44

4-12 W C SE im age ofpre-H H T specim en ......... ................ ....................... .... ...............44

4-13 WC SE image of post-HHT specimen ......... ........ ................................. 45

4-14 W C bulk carbon content .............................................................................. .............. 45

4-15 WC BSE image showing the transition zone where carbon depletion occurred ..............46

4-16 Mid-axial cross section showing WC carbon content as a function of depth...................46












-Pre-HHT Post-HHT


6000

.Zr

5000



4000



o 3000



2000


Zr
1000





-0.5 0 0.5 1 1.5 2 2.5
Energy (keV)


Figure 4-19. EDS of pre- and post-HHT of ZrC sample


3 3.5 4 4.5 5


Figure 4-20. Optical image of ZrC. A) Pre-HHT ZrC sample. B) Post-HHT ZrC sample.









4-42 TRI-C2 BSE image of pre-HHT specimen......................... ..... ...............60

4-43 TRI-C2 BSE image of post-HHT specimen................................... ...............60

4-44 TRI-C3 BSE image of pre-HHT specimen......................... ..... ...............61

4-45 TRI-C3 SE image of post-HHT specimen.............................................61

4-46 TRI-C4 BSE image of pre-HHT specimen......... ... ................................. ............... 62

4-47 TRI-C4 BSE image of post-HHT specimen................................ ...............62

4-48 TRI-C 1 pre- and post-H H T ED S.............................................. .............................. 63

4-49 TRI-C2 pre-and post-HHT ED S............... .............................................. ............... 63

4-50 TR I-C3 pre-and post-H H T ED S...................... ......... ............................... ............... 64

4-51 TR I-C 4 pre-and post-H H T E D S .............................................................. .....................64









2.1.2 Nuclear Furnace Series

The Nuclear Furnace (NF) series had a purpose similar to the Pewee series. Like the

Pewee 1, only one reactor in this series was built and operated. This was due to changes in

national priorities, which forced the Rover/NERVA program to end in 1973. It was the last

reactor built and tested during the Rover program. The NF 1 served as the test bed for two new

types of fuel forms: (U,Zr)C composite and (U,Zr)C solid solution binary carbide. This was a

44MWth reactor that operated for 109 minutes with an average exit coolant temperature of

2450K. Unlike Pewee 1, this reactor contained 49 hexagonal fuel elements. For this test, the

solid solution binary carbide fuel elements had no protective coating on the surface and showed

minimal corrosion compared to the other fuel forms. It did, however, show extensive cracking.

The composite performed better than the solid solution binary carbide in crack resistance but

corrosion from hydrogen contacting the graphite matrix continued to be a problem [5].

2.2 Literature Review

2.2.1 Hot Hydrogen Corrosion Mechanisms

A review of past studies has shown that hot hydrogen corrosion is a highly complex and

highly coupled process. That study showed that the corrosion process is a combination of the

following: (1) exposure of the fuel to hot hydrogen gas, (2) thermal and mechanical cycling and

loading, (3) radiation induced damage to the fuel matrix, and (4) annealing effect from high

temperature creep [6]. The first process is responsible for the corrosion of the microstructure.

The last three mechanisms influenced the rate at which the fuel matrix was degraded.

Hydrogen is able to penetrate the microstructure through micro-cracks and surface pores.

Depending on the local conditions such as local temperature, pressure, and composition a free

energy change or chemical reaction can occur and give rise to hydrocarbon reaction products.

Typically, reaction products occupy a greater volume than the reactants. This causes stress-

18





























Figure 4-46. TRI-C4 BSE image of pre-HHT specimen


14 v v it -



liar. =r -. ,: ,' f. .rr--' 4 ., ,
*" 1 "- J I ",- ; -' ~,"
...- ... ,, .t. _', '* 1
* X *' -Y- q- F.
",s *t ',, -. .
-. < ., .1 ." ,- P

f4 -
-Jz


-,. .. .. 7 .
,2.." ;.- L:" : ..' ^*' : ,- -.
I ,





L u 4-. .I ,i. of .









The XRD analysis was the last test to confirm an unchanged microstructure. The pre and

post-HHT ZrC samples illustrated in figure 4-27 show that the peak-to-peak positions are

unmoved, indicating the structure underwent little to no change.

4.2 Tri-Carbides

Four uranium tri-carbide compositions were tested in a hot hydrogen environment. The

carbon-to-metal ratio was fixed at 0.95. The primary goal was to examine depletion of carbon

from the microstructure, but the investigation also considered how differing the metal atom

fraction could affect the microstructure's integrity. The four tri-carbide compositions were pre-

determined by previous research [1].

Tri-carbide sample designations, compositions, and changes in density and mass are listed

in table 4-1. All of the tri-carbides displayed a mass loss as a result of the two hour hot hydrogen

test. This was expected to occur because of vaporization of material species from the surface.

Figures 4-28 through 4-31 are optical microscope images of each tri-carbide before and after

testing. No signs of surface degradation such as partial melting of individual carbide constituent

or hydrogen-induced blistering or cracking were observed.

4.2.1 Bulk Carbon Analysis

Bulk carbon analysis for each tri-carbide is given in figures 4-32 through 4-35 and table

4-2. With the exception for of TRI-C4, the other specimens' post-HHT carbon content were

within error of the pre-HHT carbon content and no change occurred from the flow of hydrogen.

The decrease of TRI-C4's carbon content can be explained by observing TRI-C4's pre-HHT

SEM image in figure 4-46. TRI-C4 showed excessive porosity, implying the specimen had not

undergone significant densification before HHT. The post-HHT samples illustrate the specimen

underwent a large degree of sintering due to exposure to the 2900K temperature. Figure 4-35

showed TRI-C4's initial carbon content at 11.775%. The high carbon content was most likely































Figure 4-40. TRI-C1 BSE image ofpre-HHT specimen


Figure 4-41. TRI-C1 BSE image ofpost-HHT specimen









analysis preparation. Qualitative elemental composition for each tri-carbide is shown in figures

4-48 through 4-51.









CHAPTER 1
INTRODUCTION

Fuel choice is a key element of nuclear reactors' use in space applications, particularly

because of concerns about fuel performance. The nuclear thermal propulsion (NTP) and some

nuclear electric propulsion (NEP) advanced space reactors must also consider the high

temperature and high power density that the fuel experiences. Nuclear reactors intended for

space applications will be required to achieve temperatures of 2500K or more. For NTP the high

temperatures ensures high specific impulse (ISP) and for NEP efficient heat rejection. High

melting point carbides or refractory carbides are capable of surviving these conditions for an

extended prototypic core lifetime, making them a suitable fuel choice.

Refractory carbides have operating temperature estimated at 3000K. This high

temperature capability makes them a preferred fuel choice, particularly when compared to other

fuel candidates such as uranium dioxide, uranium carbide, or uranium nitride. The thermal

conductivity of solid solution carbides approaches that of elemental uranium. This lowers the

thermal gradient between the fuel and propellant, thus reducing thermal stress-induced cracking.

The high thermal conductivity also allows for higher operating temperature with a smaller threat

of fuel centerline melting. Carbides also exhibit low volatility, low density, and good neutronic

properties such as good moderation due the presence of carbon and low absorption cross section.

A method for fabricating refractory carbides was established in previous research [1].

Solid solution uranium tri-carbides have been produced with the compositions (U,Zr,Nb)C,

(U,Zr,Ta)C, (U,Zr,Hf)C, and (U,Ta,Nb)C. This paper will address how microstructural integrity

changes by exposing tri-carbides to prototypic NTP/NEP temperatures. A hydrogen

environment was examined because of its role in both types of systems: in NTP systems

hydrogen is both the coolant and the propellant, while the NEP designs, such as the Pellet Bed









other the carbide component, thereby quickening the diffusion and resulting in a solid solution

tri-carbide fuel.

Figure 3-1 illustrates the hot hydrogen test equipment utilized to create the desired

conditions. Samples were placed in a chamber that is evacuated to a maximum 100millitorr then

backfilled with hydrogen in order to purge the system of oxygen and other gas impurities. This

was performed three times to minimize oxidation of the sample and chamber components. The

system was then purged and filled with hydrogen to the positive pressure, with respect to

atmospheric pressure, to ensure air could not enter the chamber. The pressure was monitored

with an Omega pressure transducer. The experiment was powered by a LEPEL Solid State RF

generator. Ultra pure hydrogen then flowed from a gas cylinder through the system for a

prototypic NTP run time. Hydrogen flow rate was measured by an OMEGA FMA-873 mass

flow controller. The temperature was monitored through the use of a MAXLINE IRCON active

control and measured by a MAXLINE IRCON two color infrared pyrometer. The pyrometer

was aligned with a viewport on the chamber to allow observations to be made. The sample was

surrounded by a tungsten susceptor that was inductively heated by a water-cooled copper coil

surrounding the tungsten susceptor. The sample was heated via radiative heat transfer from the

tungsten susceptor. Figure 3-2 shows the sample-coil assembly housed in the vacuum chamber.

Figure 3-3 shows the coil assembly with the hole drilled through the tungsten susceptor to allow

the pyrometer to measure the inner surface wall of the tungsten susceptor. This provided a more

accurate representation of the specimen temperature. A parametric temperature study, up to

2573K, was made with a C-type thermocouple to ensure accuracy of pyrometer readings.

3.3 Specimen Preparation for Analysis

After testing was complete the samples were prepared for microstructural characterization. First,

the test samples were sectioned with a LECO VC-50 diamond saw. A diamond blade was used




Full Text

PAGE 1

1 REFRACTORY CARBIDE HOT HYDROGEN ENVIRONMENT OF SPACE NUCLEAR REACTORS By BRANDON WARREN CUNNINGHAM A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLME NT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Brandon Warren Cunningham

PAGE 3

3 To the spark withi n humanity to defy conventional wisdom.

PAGE 4

4 ACKNOWLEDGMENTS I fir st want to thank my mother Tafferlon Ann Cunningham ; and my g randparents Minnie Lee and Willie Bud White Sr. They have been and continue to be instrumental in making me the person I am to day I thank the supervisory committee memb ers Dr. Edward Dugan and Dr. Luisa Amelia Dempere S pecial gratitude goes to my supervisory committee chair Dr. Samim Anghaie nuclear engineering is noteworthy of a true genius. All expe rime nts were conducted at the Applied Ultra High Te mperature Research Laboratory of the Innovative Nuclear Space Power and Propulsion Institute (INSPI) at the Universit y of Florida. Analysis was performed at the Major Analytical Instrumentation Center at the University of Florida

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIG URES ................................ ................................ ................................ ......................... 8 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 INTRODUCTI ON ................................ ................................ ................................ .................. 14 2 BACKGROUND AND LITERATURE REVIEW ................................ ................................ 16 2.1 Historical Relevance for Fuel Development ................................ ................................ 16 2.1.1 Pewee Series ................................ ................................ ................................ ..... 17 2.1.2 Nuclear Furnace Series ................................ ................................ ..................... 18 2.2 Literature Review ................................ ................................ ................................ ......... 18 2.2.1 Hot Hydrogen Corrosion Mechanisms ................................ ............................. 18 2.2.2 Free Energy of Formation Calculations ................................ ............................ 19 3 EXPERIMENT METHODS AND PROCEDURES ................................ .............................. 24 3.1 Mass Flow Determination ................................ ................................ ............................ 24 3.2 Equipment and Testing Procedures ................................ ................................ ............. 24 3.3 Specimen Preparation for Analysis ................................ ................................ .............. 25 3.4 Analysis Methodology ................................ ................................ ................................ 27 4 RESUL TS AND DISCUSSION ................................ ................................ ............................. 31 4.1 Mono Carbides ................................ ................................ ................................ ............ 31 4.1.1 Tantalum Carbide ................................ ................................ ............................. 31 4.1.2 Tungsten Carbide ................................ ................................ .............................. 32 4.1.3 Zirconium Carbide ................................ ................................ ............................ 34 4.2 Tri Carbides ................................ ................................ ................................ ................. 35 4.2.1 Bulk Carbon Analysis ................................ ................................ ....................... 35 4.2.2 XRD Analysis ................................ ................................ ................................ ... 36 4.2.3 SEM/EDS Analysis ................................ ................................ .......................... 36 5 CONCLUSIONS AND RECOMMENDATIONS ................................ ................................ 65

PAGE 6

6 LIST OF REFERENCES ................................ ................................ ................................ ............... 67 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 69

PAGE 7

7 LIST OF TABLES Table page 3 1 Average testing conditions for mono carbide samples ................................ ...................... 28 3 2 Average testin g conditions for tri carbide samples ................................ ........................... 28 4 1 Uranium tri carbide results ................................ ................................ ................................ 38 4 2 Uranium tri carbide pre and post HHT carbon cont ent ................................ .................... 38

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Four fuel forms used in Rover/NERVA designs (Matthews et al., 1991) .......................... 21 2 2 Pseudo binary phase diagram of (U 0.1 ,Zr 0.9 )C x (Lyon, 1973) ................................ ............ 22 2 3 Standard free energy of formation of possible chemical reactions of mono carbide s in a hydrogen environment [8,9,10] ................................ ................................ ....................... 23 3 1 Hot hydrogen test chamber ................................ ................................ ................................ 29 3 2 Top view of sample/tungsten susceptor/ind uctive coil configuration ................................ 29 3 3 Copper cooled induction coil with tungsten susceptor assembly ................................ ....... 30 4 1 EDS of pre and post HHT TaC sample ................................ ................................ ............. 39 4 2 Optical image of TaC. A) Pre HHT TaC sample. B) Post HHT TaC sample ................. 39 4 3 TaC BSE image showing post HHT cracking ................................ ................................ .... 40 4 4 TaC SE image showing fracture zone ................................ ................................ ................ 40 4 5 TaC SE image of pre HHT specimen ................................ ................................ ................. 41 4 6 TaC SE image of post HHT specimen ................................ ................................ ............... 41 4 7 TaC bulk carbon content ................................ ................................ ................................ ..... 42 4 8 Mid axial cross section showing TaC carbon content as a function of depth .................... 42 4 9 TaC pre and post HHT XRD spectrum ................................ ................................ ............. 43 4 10 EDS of pre and post HHT of WC sample ................................ ................................ ......... 43 4 11 Optical image of WC. A) Pre HHT WC sample. B) Post HHT WC sample ................... 44 4 12 WC SE image of pre HHT specimen ................................ ................................ ................. 44 4 13 WC SE image of post HHT specimen ................................ ................................ ................ 45 4 14 WC bulk carbon con tent ................................ ................................ ................................ ..... 45 4 15 WC BSE image showing the transition zone where carbon depletion occurred ................ 46 4 16 Mid axial cross section show ing WC carbon content as a function of depth ..................... 46

PAGE 9

9 4 17 Mid axial cross section of post HHT showing WC transition zone of carbon content ...... 47 4 18 WC pre and post HHT XRD spectrum ................................ ................................ ............. 47 4 19 EDS of pre and post HHT of ZrC sample ................................ ................................ ......... 48 4 20 Optical image of ZrC. A) Pre HHT ZrC sample. B) Post HHT ZrC sample. .................. 48 4 21 ZrC SE image of pre HHT specimen ................................ ................................ ................. 49 4 22 ZrC SE image of post HHT specimen ................................ ................................ ................ 49 4 23 ZrC BSE image of pre HHT specimen ................................ ................................ ............... 50 4 24 ZrC BSE image of post HHT specimen ................................ ................................ ............. 50 4 25 ZrC bulk carbon content. ................................ ................................ ................................ .... 51 4 26 Mid axial cross section showing ZrC carbon content as a function of depth. .................... 51 4 27 Z rC pre and post HHT XRD spectrum ................................ ................................ ............. 52 4 28 Optical image of TRI C1. A) TRI C1 pre HHT. B) TRI C1 post HHT ......................... 53 4 29 Optical image of TRI C2. A) TRI C2 pre HHT. B) TRI C2 post HHT .......................... 53 4 30 Optical image of TRI C3. A) TRI C3 pre HHT. B) TRI C3 post HHT .......................... 54 4 31 Optical image of TRI C4. A) TRI C4 pre HHT. B) TRI C4 post HHT .......................... 54 4 32 TRI C1 pre and post HHT bulk carbon content ................................ ................................ 55 4 33 TRI C2 pre and post HHT bulk carbon content ................................ ................................ 55 4 34 TRI C3 pre and post HHT bulk carbon content ................................ ................................ 56 4 35 TRI C4 pre and post HHT bulk carbon content ................................ ................................ 56 4 36 TRI C1 XRD spectrum ................................ ................................ ................................ ....... 57 4 37 TRI C2 XRD spectrum ................................ ................................ ................................ ....... 57 4 38 TRI C3 XRD spectrum ................................ ................................ ................................ ....... 58 4 39 TRI C4 XRD spectrum ................................ ................................ ................................ ....... 58 4 40 TRI C1 BSE image of pre HHT specimen ................................ ................................ ......... 59 4 41 TRI C1 BSE image of post HHT specimen ................................ ................................ ....... 59

PAGE 10

10 4 42 TRI C2 BSE image of pre HHT specimen ................................ ................................ ......... 60 4 43 TRI C2 BSE image of post HHT specimen ................................ ................................ ....... 60 4 44 TRI C3 BSE image of pre HHT specimen ................................ ................................ ......... 61 4 45 TRI C3 SE image of post HHT specimen ................................ ................................ .......... 61 4 46 TRI C4 BSE image of pre HHT specimen ................................ ................................ ......... 62 4 47 TRI C4 BSE image of post HHT specimen ................................ ................................ ....... 62 4 48 TRI C1 pre and post HHT EDS ................................ ................................ ........................ 63 4 49 TRI C2 pre and post HHT EDS. ................................ ................................ ........................ 63 4 50 TRI C3 pre and post HHT EDS. ................................ ................................ ........................ 64 4 51 TRI C4 pre and post HHT EDS ................................ ................................ ......................... 64

PAGE 11

11 LIST OF ABBREVIATIONS BSE Back s catter ed Electron C wt% Carbon Weight percent EDS Energy Dispersive Spectrum EMPA Electron Micro Probe Analysis HHT Hot Hydrogen Testing HIP Hot Isostatic Press INSPI Innovati ve Nuclear Space Power and Propulsion Institute ISP Specific Impulse LANL Los Alamos National Laboratory MW th Megawatt thermal NASA MSFC National Aeronautics and Space Administration Marshall Space Flight Center NEP Nuclear Electric Propulsion NERVA Nu clear Engine Rocket Vehicle Application NF Nuclear Furnace NTP Nuclear Thermal Propulsion PyC Pyrolitic Carbon SE Secondary Electron SEM Scanning Electron Microscop y TaC Tantalum Carbide WC Tungsten Carbide XRD X Ray Diffraction ZrC Zirconium Carb ide f Standard free energy of formation

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science NTEGRITY IN HOT HYDROGEN ENVIRONMENT OF SPACE NUCLEAR REACTORS By Brandon Warren Cunningham August 2008 Chair: Samim Anghaie Major: Nuclear and Radiological Engineering This project required the testing of three mono carbide s to investigate which carbi de has the least degradation of microstructural properties when exposed to a hot hydrogen environment This was performed in order to determine the usability of various carbides in a base mixture The purpose of this research was to investigate carbides as candidate fuel matrix for nuclear thermal propulsion. Tantalum carbide (TaC), tungsten carbide (WC), and zirconium carbide (ZrC) were all tested in a hot hydrogen environment at an average temperature of 2775K. Each study tracked the carbon content, c orrosion, density, hydrogen embrittlement, and phase changes. The results show that some carbides experience d changes in all of the aforementioned properties while others experienced a combination of changes or no noticeable degradation. The second sta ge of the research tested solid solution uranium bearing tri carbides in a hot hydrogen environment. The tri carbides included the compositions (U 0.1 Zr 0.58 Nb 0.32 )C 0.95 (U 0.1 Zr 0.68 Nb 0.22 )C 0.95 (U 0.1 Zr 0.77 Nb 0.13 )C 0.95 and (U 0.05 Zr 0.62 Nb 0.3 3 )C 0.95 All tri carbides were exposed to temperatures above 2900K for two hours. Scanning electron microscopy, bulk carbon analysis, X ray diffraction, density measurements and mass measurements were used to characterize each specimen Analysis showed that differences in metal composition ha d no

PAGE 13

13 noticeable effect on final fuel integrity. Carbon analysis electron microscopy, and diffraction analysis showed th at tri carbides display ed little to no change in micro structure due to exposure to a hot hydro gen environment

PAGE 14

14 CHAPTER 1 I NTRODUCTION Fuel choice is a key element of nuclear reactors use in space applications, particularly because of concerns about fuel performance. The nuclear thermal propulsion ( NTP ) and some nuclear electric propulsion ( NEP ) advanced space reactors must also consider the high temperature and high power density that the fuel experiences Nuclear reactors intended for space applications will be required to achieve temperatures of 2500K or more. For NTP the high temperatures e nsures high specific impulse ( ISP ) and for NEP efficient heat rejection. High melting point carbides or r efractory carbide s are capable of surviving these conditions for an extended prototypic core lifetime making them a suitable fuel choice Refractor y carbide s have operating temperature estimated at 3000K. This high temperature capa bility makes them a preferred fuel choice, particularly when compared to other fuel candidates such as uranium dioxide uranium carbide, or uranium nitride. The thermal c onductivity of solid solution carbides approaches that of elemental uranium This lowers the thermal gradient between the fuel and propellant thus reducing thermal stress induced cracking. The high thermal conductivity also allows for higher operating t e mperature with a smaller threat of fuel centerline melting. Carbides also exhibit low vol at ility, low density, and good neutronic properties such as good moderation due the presence of carbon and low absorption cross section A method for fabricating r efractory carbide s was established in previous research [1] Solid solution uranium tri carbide s have been produced with the compositions (U,Zr,Nb)C, (U,Zr,Ta)C, (U,Zr,Hf)C, and (U,Ta,Nb)C. This paper will address how microstructural integrity change s by exposing tri carbides to prototypic NTP/NEP temperatures A hydrogen environment was examined because of its role in both types of systems: in NTP systems hydrogen is both the coolan t and the propellant while the NEP designs such as the Pellet Bed

PAGE 15

15 Reac tor concept or Neptune reactor concept it acts as a coolant Uranium tri carbide was chosen for test ing because preliminary results show ed that tri carbides have higher melting temperatures and better resistance to hot hydrogen corrosion than the bi nary carbide fue l (U,Zr)C [2] Testing materials in a high temperature hydrogen environment is rare because most operating conditions do not warrant its use As a result, background information and literature is limited. The main concern with NTP is hot hydrog en corrosion. H ydrogen at high temperatures can erode carbon from the matrix or reac t and corrode the fuel substrate to form methane or other hydrocarbons Studies of previous NTP programs and case studies have shown that the hot hydrogen corrosion mecha ni sms are a highly coupled and highly complex process Research shows that material properties are depende nt on maintaining stoichiometry [3] Changes in the carbon to metal ratio can lead to changes in the liqui dus, changes in phase ori entation fatigue resistance and toughness. A quantification of carbon to metal ratio with exposure time to temperature is needed to ensure properties such as melting point, thermal conductivity, and co nsistent thermal expansion are coherent throughout t lifetime E vidence of hot hydrogen corrosion can show as hydrogen induced blistering, hot embrittlement, mass loss and hydride formation.

PAGE 16

16 CHAPTER 2 BACKGROUND AND LITER ATURE REVIEW 2.1 Historical Relevance for Fuel Development Nucl ear reactors for rocket applications w ere first ground tested during the Rover program The main purpose of the Rover program was to develop nuclear powere d rockets for ballistic missile defense. When the need for nuclear rocket applications diminished d ue to the efficiency of chemical rockets, Los Alamos National Laboratory ( LANL ) contracted with Westinghouse Electric Corporation to create a research and development program within the original framework of the Rover program The new program, the N uclear E ngine for R ocket Vehicle A pp lications (NERVA), was developed for space purpose s The NERVA program was charged not only with reactor design, but also to achieve the highest possible propellant temperature. The intent was a completed rocket design with maximize d ISP while maintaining low weight. Equation 1 defines ISP as proportional to the square root of the fluid temperature over the fluid molecular weight [2] Hydrogen is the de facto coolant for NTP because its molecular weight is the lowest and hig hest specific heat of all elements All parameters being equal, the temperature of the coolant dictates the efficiency of the rocket design. Therefore, fuel development was rigorously pursued in order to meet this goal Throughout the course of the Rover program several reactors series with varying fuel forms were investigate d Figure 2 1 illustrates the fuels used during the Rover /NERVA program. Nuclear reactor cores developed during these program s generally operated with temperatures around 2500K and pressures of 3MPa. The series that is most important for this research, the Pewee and Nuclear Furnace families, focused on alternative fuels, higher fuel temperature and fuel performance

PAGE 17

17 Eqn. 1 A = performance factor related t o thermo physical properties of the propellant C f = thrust coefficient, which is a function of nozzle parameters T ch = chamber temperature (K) M c = molecular weight of propellant F = thrust (N) m = mass flow rate of propellant (kg/s) 2.1.1 Pewee Series Th e P ewee series was intended to be a family of small inexpensive test reactors use d for high temperature fuels demonstration : only one reactor was eventually built and demonstrated. In 1968 the Pewee 1 ran successfully at 503MW th for 40 min ute s with an exi t coolant temperature of 2550K. The peak fuel temperature was measured at 2750K. It demonstrated the highest power for a reactor of its s ize with an average power density of and a peak power density of The calculated ISP in vacuum was 845s. Pewee contained 402 hexagon al f uel elements loaded with pyrolytic graphite coated UC 2 particles dispersed throughout a graphite matrix. The fuel elements were coated with NbC, ZrC, and Mb as an overcoat I nspection of the fuel elements after testing s howed that many of the elements exhibited cracked regions or external corrosion Fractures were due to thermal mechanical stresses but corrosion was attributed to the fuel elements graphite ma trix being subjected to a hot hydrogen environment [ 4 ]

PAGE 18

18 2.1 .2 N uclear F urnace Series The Nuclear Furnace (NF) series had a purpose similar to the Pewee series L ike the Pewee 1 only one reacto r in this series was built and operated This was due to chan ges in national priorities which forced the Rover/NERVA program to end in 1973. It was the last reactor built and tested during the Rover program. The NF 1 served as the test bed for two new types of fuel forms : (U,Zr)C composite and (U,Zr)C solid solut ion binary carbide This was a 44MW th reactor that operated for 109 min utes with an average exit coolant temperature of 2450K Unlike Pewee 1, t his reactor contained 49 hexagonal fuel elements For this test the solid solution binary carbide fuel eleme nts had no protective coating on the surface and showed minima l corrosion compared to the other fuel form s It did, however show extensive cracking The composite performed better than the solid solution binary carbide in crack resistance but corrosion from hydrogen contacting t he graphite matrix continued to be a problem [5 ] 2.2 Literature Review 2.2.1 Hot Hydrogen Corrosion Mechanisms A review of past studies has shown that hot hydrogen corrosion is a highly complex and highly coupled process That s tudy showed that t he corrosion process is a combination of the following: (1) exposure of the fuel to hot hydrogen gas, (2) thermal and mechanical cycling and loading (3) radiation induced damage to the fuel matrix, and (4) annealing effect from high temp erature creep [6 ] The first process is responsible for the corrosion of the microstructure. The last three mechanism s influence d the rate at which the fuel matrix was degraded Hydrogen is able to penetrate the microstructure through micro cracks and su rface pores. Depending on the local conditions such as local temperature, pressure, and composition a free energy change or chemical reaction can occur and giv e rise to hydrocarbon reaction products Typically reaction product s occupy a greater volume t han the reactants. This cause s stress

PAGE 19

19 corrosion cracking which consequently creates a large r surface area and deeper route for the hot hydrogen to attack. The corrosion induced cracks and pores weaken the microstructure and allow for surface erosion. T he depletion of carbon from the fuel surface leads to changes in local composition which alters local material properties In solid solution uranium bi carbides it has been shown that material properties depend heavily on the carbon to metal ratio. Figu re 2 2 shows a phase diagram of a pseudo binary carbide (U 0.1 Zr 0.9 )C x In this composition the carbon to metal ratio is limited to 0.92 0.96 before a so lid fuel transforms to a mixture of liquid/ solid solution carbide Allowing this to occur will ultim ate ly lead to fuel failure. Fuel irradiation enhances the rate of hot hydrogen corrosion Exposure of the microstructure to radiation, particularly fast neutrons, create s lattice dislocations and point defects The localized defect effectively reduces th e local thermal conductivity which produc es a thermal gradient with the surrounding lattice sites The defect s also reduce local ductility This combined with the reduction in thermal cond uctivity produces thermal stress induced cracking that expos es m ore surface area to hot hydrogen attack. Creep i s a mechanism that does not enhance the hot hydrogen corrosion rate, unlike the other two processes. The operating temperatures of NTP fuel accelerate the creep process. Creep will serve to a nneal the loc alized high stress zones deterring the formation of crack growth and closing micro cracks within the microstructure. 2.2.2 Free Energy of Formation Calculations Binary carbides and ternary carbides are material s whose full use ha s yet to be investigated As a result, there is no tabulated data on the free energy of formation to determine the stability The f ree energy of formation for the mono carbide s was calculated to determine the system s stability in a hydrogen environment. An approximation of

PAGE 20

20 ternary carbide s ystem stability was made by observing the G ibbs potential for mono carbides since uranium tri carbide is a solid sol ution of (U,Zr,X)C where X is a metal. The second law of thermodynamics was used t o determine whether the mechanism for microstructural damage to mono carbide s was predominantly corrosion or erosion The Standard f ) gives the work released or consume d by a chemical reaction at f is posit ive the chemical reaction is non spontaneous f is negative the chemical reaction f means that the species reaction cannot take place unde r the given conditions without the f calculated [7] G f was calculated for several chemical reactions that are possible in a hydrogen environment neglecting the presence of trace elements such as oxygen in the testing system The dissociation energ y of diatomic hydrogen is 4.5eV. T his is higher than the thermal energy of the system, 2 7 0 0 K or 0.2 32 eV For simplification, o nly atomic hydrogen would be consider ed because of its chemical volatility Figure 2 3 displays the Gibbs free energy plot for several reactions. With the e xcept ion of one chemical re action each of the other reaction s are not possible at the temperatures of interest without the addition of more work. This implies that the main mechanism for microstructural damage is erosion rather than corrosion. The one chemical reaction that is spo 4 which readily occur s at temperatures beneath 445K or its equilibrium temperature. Th is is a reaction that is plausible for corrosion to occur and carbon loss to be significant. Using this principle the mono carbide, WC should ex perience a noticeable loss in carbon content. Since erosion is the mechanism for damaging the TaC and ZrC matrix those compounds could exhibit a uniform loss of both metal and carbon.

PAGE 21

21 Figure 2 1. Four fuel forms used in Rover/NERVA designs (Matthews e t al., 1991)

PAGE 22

22 Figure 2 2. Pseudo binary p h ase d iagram of (U 0.1 ,Zr 0.9 )C x (Lyon 1973)

PAGE 23

23 Figure 2 3 Standard free energy of formation of possible chemical reactions of mono carbides in a hydrogen environment [ 8,9,10 ]

PAGE 24

24 CH APTER 3 EXPERIMENT METHODS A ND PROCEDURES 3.1 Mass Flow Determination A linear relationship was used t o ensure that the samples tested in this experiment experienced a relative mass flow rate similar to a nuclear space reactor The Pewee reactor from the R over/NERVA program was used as a template. The Pewee reactor used a total hydrogen flow rate of with cooling the fuel The total available coolant area of 858,929cm 2 was base d on dimension s of t he core [4] This gives a mas s flow per square centimeter of The mono carbide and tri carbide specimen tested in this experiment had a n available coolant area of approximately 3.5cm 2 resulting in a n equivalent mass fl ow rate of The dimensions of each specimen vary slightly making 3.5cm 2 the estimated sample surface the coolant contacts Chamber pressure, temper ature, and mass flow rate were recorded every 10 min utes to calculate a nd average testing conditions. All average testing conditions are listed in table s 3 1 and 3 2. 3 .2 Equipment and Testing Procedures Two series of samples were tested. Mono carbide samples produced at National Aeronautic s and Space A dministration all Space Flight Center ( NASA MSFC ) and u ranium tri carbides produced at and Propuls ion Institute ( INSPI ) The mono carbide s were fabricated by hot Isostatic press (HIP) Tri carbides were fabricated by an induction heated uni axial hot press liqui d phase sintering method [1] During this process a homogeneous mixture of three carbides are cold pressed and heated to a temperature above the lowest melting point carbide constituent in this case uraniu m carbide. The liquid uranium carbide fills in porous areas and contacts the grain boundary from

PAGE 25

25 o ther the carbide component, thereby quickening the diffusion and resulting in a solid solution tri carbide fuel. Figure 3 1 illustrates the hot hydrogen test equipment utilized to create the desired conditions. Samples were placed in a chamber that is evacuated to a maximum 1 00millitorr then backfilled with hydrogen in order to purge the system of oxygen and other gas impurities This was performed three tim es t o minimize oxidation of the sample and chamber components. The system was then purged and filled with hydrogen to the positive pressure with respect to atmospheric pressure, to ensure air could not enter the chamber. The pressure was monitored with an Omega pressure transducer. The experiment was powered by a LEPEL Solid State RF generator Ultra pure h ydrogen then flowed from a gas cylinder through the system for a prototypic NTP run time Hydrogen flow rate was measure d by an OMEGA FMA 873 mass flow controller. The temperature wa s monitored through the use of a MAX LINE IRCON active control and measured by a MAXLINE IRCON two color infrared pyrometer. The pyrometer wa s aligned with a viewport on the chambe r to allow observations to be made The sample wa s surrounded by a tungsten sus ceptor that was inductively heated by a water cooled copper coil surrounding the tungsten susceptor. The sample wa s heated via radiative heat transfer from the tungsten susceptor. Figure 3 2 shows the sample coil a ssembly housed in the vacuum chamber. Figure 3 3 shows the coil assembly with the hole drilled through the tungsten susceptor to allow the pyrometer to measure the inner surface wall of the tungsten susceptor. This provided a more accurate representation of the specimen temperature. A parametric temperature study, up to 2573K, was made with a C type thermocouple to ensure accuracy of pyrometer readings. 3.3 Specimen Preparation for Analysis After testing was complete the sample s were prepar ed for m icrost ructural characterization. First, the test samples were sectioned with a LECO VC 50 diamond saw. A diamond blade was used

PAGE 26

26 to account for the hard ness of carbides. Individua l sections were taken for X ray Diffraction ( XRD ) bulk carbon analysis, Electro n MicroProbe Analysis ( EMPA ) and Scanning Electron Microscope ( SEM ) analysis All specimens were ultrasonic ally cleaned and degassed in distilled water for 60 minutes with a Bransonic B3 R. All specimens were dried in a desiccator for a minimum of 24 h ours before the analysis was conducted a. XRD Preparation i. Specimen s were either pulverized to a powder in an alumina crucible or sectioned to approximately 0.5mm thickness and mounted on glass slides using a Amyl acetate and Collodian solution b. Bulk Carbon Analysis i. Specimens prepared for bulk carbon analysis were pulverized to a powder in an alumina crucible ii. Three to five batches weighing 0.25g f rom each sample composition w ere placed in a zirconia crucible. High purity accelerator was added to ensure sam ples could be inductively heated and ignited. A mixture of 1.8 g of copper and 1.2g of iron were the amounts recommended by the equipment operation s manual The iron carbon impurity content was 8PPM. iii. Samples were placed in carbon analyzer for combustion analysis. The carbon determinator was calibrated using a tungst en carbide standard with 6.18wt% C c. SEM and EMPA P reparation i. Sections for SEM and EMPA were mounted in phenol ring molds using a 1:5 ratio LECO hardener to LECO epoxy resin respectively ii. LECO Spectrum System 1000 with semi automatic head s w as used for rough polishing mounted specimens from 180grit to 1200grit. iii. The final polishing steps decreased Polishing was performed by hand. iv. Section s intend ed for SEM were acid etched for one minute with HNO 3 /HCl/H 2 SO 4 solution v. Sections intended for EMPA was not acid etched.

PAGE 27

27 3. 4 Analysis Methodology The following variable s were examined for pre and post hot hydrogen testing (HHT): overall carbon concentration, carbon and metal concentration as a function of depth, density, and phase and crystallographic changes S everal visual observations such as grain growth, inter and intra granular corrosion or erosion, spallation, cracking, swelling, and hydrogen embrittlement were also made The pre and post HHT density was measured by immersion testing. A LECO WC 200 carbon determinator was used to combust specimens t o investiga te the bulk carbon concentration A carbon profile of each specimen as a function of depth was characterized by EMPA The EMPA instrument used was a JEOL SUPERPROBE 733 Electron Probe. Topographical and compositional contrast and elemental composition w ere inspected with a JEOL JSM 6400 SEM SEM was conducted to make most visual observations. Using XRD phase and crystallographic properties were analyzed with Philips APD 3720 X Ray Diffractometer and reference spectra database [1 1 ] Specimen weight an d dimensions were measured with the Sartorius R180D analytic balance with an accuracy of 0.00001g and a Mitutoyo Absolute Digimatic Ca li per (CD

PAGE 28

28 Table 3 1. Average testing conditions for mono carbide sample s TaC WC ZrC Mass Flow Rate 15.58 15.19 15.74 Temperature 2762 2779 2776 Kelvin Chamber Pressure 968 969 970 torr Run Time 2 2 1.167 hr *The ZrC sample test was conducted for a shorter time because of equipment constraints. Table 3 2. Average testing conditions for tri carbide sample s (U 0.1 Zr 0.58 Nb 0.32 )C 0.95 (U 0.1 Zr 0.68 Nb 0.22 )C 0.95 (U 0.1 Zr 0.77 Nb 0.13 )C 0.95 (U 0.05 Zr 0.62 Nb 0.33 )C 0.95 Mass Flow Rate 15.81 14.92 14.92 14.92 Temperature 2926 2903 2903 2903 kelvin Chamber Pressure 982 964 964 964 torr Run Time 2 2 2 2 hr

PAGE 29

29 Figure 3 1. Hot hydrogen test chamber Figure 3 2 Top view of sample/tungsten susceptor/inductive coil configuration

PAGE 30

30 Figure 3 3 Copper cooled induction coil with tungsten susceptor assembly

PAGE 31

31 CHAPTER 4 RESULTS AND DISCUSSI ON 4.1 Mono C arbide s 4.1.1 Tantalum Carbide To begin the analysis an energy dispersive s pectr um (EDS ) w as taken to verify that the sample contains the given elements. Figure 4 1 shows the EDS both pre and post HHT. The post HHT spectrum was taken to verify the re was no conta mination ato mic numbers higher than carbon were introduced to the sample after HHT. Figure 4 2 gives a view of the pre and post HHT TaC samples. Compared to the pre HHT sample the post HHT shows extensive cracking on the surface. A cros s section was taken through the post sample and subsurface cracking was also present. Figure s 4 3 and 4 4 show the extent of cracking through a mid axial cross section. Figure s 4 5 and 4 6 show that the grain structure was unaffected by the presence of h ydrogen. Neither grain growth nor grain degradation occurred, but the grain area adjacent to the cracks was fracture d Fractures could be attributed to fast heat up rate. TaC has a theoretical carbon weight percent (C wt % ) of 6.224wt%. Figure 4 7 shows the pre HHT C wt% at 8.136wt% and post HHT at 7.975wt%, both of which are hyper stoichiometric. Statistically there was no change in carbon content. To determine if that small difference in the mean C wt% for the pre and post HHT samples is due to the depletion of carbon at the surface from hydrogen flow, an E MP A was per formed. The resul ts of the EMPA shown in figure 4 8 illustrates that the pre and post HHT samples show no change i n carbon represents the center of the sample or 3mm from the edge. The fluctuation in the EMPA profile is caused by surface roughness The surface roughness scatter s t he incident electrons in directions that exaggerate or decrease the counts in the solid angle of the detector

PAGE 32

32 The immersion density test show ed that for the TaC pre HHT sample the density was The post HHT sample had yielding a percent change of 4.836%. The reference theoretical density for TaC is making the pre HHT sample 98.98% dense and the post HHT sample 94.19% dense after a two hour run time. Since pores are often the sites for crack initiation it is possible that pores are the cause of the extensive cracking throughout the TaC sample. The XRD analysis for the pre and post HHT show ed that t he structure did not undergo a phase change. Figure 4 9 displays the results. If an y of the reactions in figure 2 3 would have occurred, peaks indicating elemental tantalum would have been present 4.1. 2 Tungsten Carbide The EDS displayed in figure 4 10 indicates that the elemental compositions for the pre and post HHT samples are consis tent. The other analysis for the post HHT sample shows a large difference in properties from the pre HHT. The blistering on the surface in figure 4 11 b indicates the formation of a liquid or a gas. Assuming there were trace amounts of oxygen in the syst em and the formation of tungsten oxide was negligible, the blistering on the surface could be attributed to trapped methane gas. The post HHT WC sample show s that the degradation occurred at the grain boundaries. An explanation for the blistering can att ributed to the Gibbs potential calculated for WC. H ydrogen reacted with carbon in the WC matrix forming methane or hydrocarbons. That methane gas then diffuses though the path of least resistance or the boundaries to the surface. Th e gas that did not es cape was trapped under the surface to blister. Significant grain growth occurred in the WC sample. Figure 4 12 and figure 4 13 show the pre HHT and post HHT grain structure respectively. T his can be attributed to conducting the testing near the melting point of tungsten carbide (3143K).

PAGE 33

33 Figure 4 1 4 indicates t h at the bulk carbon content for the pre HHT sample was 12.7% while the post HHT sample had a C wt% of 5.928 Since pre and post HHT carbon content are outside of their statistical error then deplet ion of carbon from the microstructure occurred with certainty. The large difference in the pre and post HHT carbon content was due to carbon depletion at the surface of the WC in the post HHT sample. Figure 4 15 shows a BSE image of the WC post HHT. Th e brighter edge region indicates there was higher tungsten content at the edge. The depletion of carbon at the edge was verified using EMPA Figure 4 1 6 show s that the post a point center of the sample indicates there is a diffusion gradient of carbon in the post HHT sample. Figure 4 1 7 HH zone returns to the carbon concentration of the pre HHT sample. This gives a carb on depletion depth rate approximately Notice the difference between the pre HHT bulk 12.7C wt% and EMPA HHT avera ge trend which shows a lower C wt%. Figure 4 15 shows dark spherical regions rich in carbon that would have be en include d in the bulk carbon analysis but not the EMPA result. Therefore the C wt% for the WC matrix follows the EMPA trend, average 7.75wt%, but the wt% is 12.7%. These rich carbon regions were probably introduced during the sample s fabrication as contamination There is a large difference in the pre and post HHT density. The pre HHT density is while the post HHT sample is Th is gives a percent difference of 10.91%

PAGE 34

34 The XRD spectrum shows that there were significant changes in phase. The pre HHT sample in figure 14 1 8 illustrates that only the WC phase is present. The post HHT sa mple reveals the presence of two phases: WC and W 2 C. The presence of the W 2 C is evidence that the carbon to metal ratio has decreased due to depletion of carbon from the microstructure surface 4.1. 3 Zirconium Carbide The EDS for the ZrC pre and post HHT samples are displayed figure 4 1 9 Th e results show that both sample s composition is consist ent throughout the experiment Figure 4 20 a and 4 20 b show s that on the surface the post HHT sample displays little to no change s when compared to the pre HHT s ample. There is a slight discoloration caused by hot hydrogen etching the carbonaceous or loose mat erial on the surface. Figure s 4 21 and 4 22 show that qualitatively the microstructure was unchanged. The BSE images in 4 23 and 4 24 show no compositiona l differences or any depleted carbon regions due to hydrogen flow. A bulk carbon analysis confirmed the qualitative result of an unchanged post HH T ZrC microstructure. Figure 4 25 illustrates the pre HHT C wt% valued at 16.450wt% and the post HHT C wt% va lued at 16.775wt%. Though the post HHT sample is higher, the values are within close statistical error. The theoretical ZrC carbon content is 1 1.634wt%. The EMPA in figure 4 26 shows that there was no depletion of carbon at the surface of the sample. Th e variability in data points is caused by surface roughness. Changes in de nsity could reveal the presence of hydrogen embrittlement. The pre HHT while the post The change in density from the control sample is 0.197%. Given the precision of the immersion testing the values of the pre and post HHT density are statistically the same.

PAGE 35

35 The XRD analysis was the last test to confirm an unchanged microstructure. Th e pre and post HHT ZrC samples illustrated in figure 4 27 show that the peak to peak positions are unmoved indicating the s tructure underwent little to no change. 4.2 Tri Carbides Four uranium tri carbide composition s wer e test ed in a hot hydrogen environ ment. The carbon to metal ratio was fixed at 0.95 The primary goal was to examine depletion of carbon from the microstructure, but the investigation also considered how differing the metal atom fraction could T he four tri carbide compositions were pre determined by previous research [1] Tri carbide sample designation s composition s, and changes in density and mass are listed in table 4 1. All of the tri carbides displayed a mass l oss as a result of the two hour hot hydrogen test This was expected to occur because of vaporization of material species from the surface Figures 4 28 through 4 31 are optical microscope images of each tri carbide before and after testing. No signs of surface degradation such as pa rtial melting of individual carbide constituent or hydrogen induced blistering or cracking w ere observed 4.2. 1 Bulk Carbon Analysis Bulk carbon analysis for each tri carbide is given in figures 4 32 through 4 35 and table 4 2 With the exception for o f TRI C4, t he post HHT carbon content w ere within error of the pre HHT carbon content and no change occurred from the flow of hydrogen The decrease of TRI carbon content can be explain ed by observing TRI HHT SEM image in fi gure 4 46 TRI C4 showed excessive porosity, implying the specimen had not undergone significant densification before HHT The post HHT samples illustrate the specimen underwent a large degree of sintering due to exposure to the 2900K temperature. Figur e 4 35 showed TRI at 11.775 % The high carbon content was most likely

PAGE 36

36 due to carbonaceous material contaminating the surface and filling the pores Samples TRI C1, TR I C2, and TRI C3 showed no measurable decrease in carbon con tent implying little to no erosion or corrosion of carbon from the tri carbide matrix occurred The bulk carbon analysis demonstrated that the tri 4.2.2 XRD Analysis Another measure of changes in t he microstructure would have been revealed through XRD. None of the post HHT diffraction patterns in figures 4 36 through 4 39 show ed the presence of different phase s from that of the pre HHT sample diffraction pattern The structure was c onsider ed to be a solid solution because the diffraction pattern did not exhibit the presence of segregated carbide constituents. Observing the XRD pattern of NbC, ZrC, and UC/UC 2 powders from a previous research [1] the diffraction pattern of the tri ca rbides verifie d that a solid solution of (U,Zr,Nb)C exist s in both th e pre and post HHT specimens N o changes in phase or crystallographic orientation occurred as a result of the two hour test at 2900K. 4.2.3 SEM/EDS Analysis Using SEM c omposition al con trast and topographical images were obtained. Figures 4 40 through 4 47 are images of the specimen microstructure. Heterogeneity of Zr and Nb in the microst ructure could not be ascertained from BSE images due to the proximity of their atomic numbers. It can be inferred that had the pre and post HHT XRD patterns displayed Zr C and Nb C spectra, the two elements would have been segregat ed in the microstructure Knowing that this was not the case, Zr and Nb could be assumed to have a homogeneous distribut ion. Based on compositional contrast images, a ll the tri carbide s displayed no regions of depleted carbon or a gradient of uranium from the surface. In addition to SEM images, EDS was taken of each pre and post HHT sample to ensure no contamination was introduced to the sample during testing or

PAGE 37

37 analysis preparation. Qualitative elemental composition for each tri carbide is shown in figures 4 48 through 4 51.

PAGE 38

38 Table 4 1. Uranium tri carbide results Sample Designation Tri carbide composition (atom frac tion) Pre HHT Density (g/cm 3 ) Post HHT Density (g/cm 3 ) Relative density difference Pre HHT Mass (g) Post HHT Mass (g) Mass Difference (g) Percentage of mass loss TRI C1 (U 0.1 Zr 0.58 ,Nb 0.32 )C 0.95 6.1066 5.7567 5.73% 2.24545 2.11287 0.13258* 5.9% TRI C 2 (U 0.1 Zr 0.68 ,Nb 0.22 )C 0.95 6.5306 6.1129 6.40% 2.53249 2.43116 0.10133* 4.0% TRI C3 (U 0.1 Zr 0.77 ,Nb 0.13 )C 0.95 6.2807 6.4215 +2.24% 2.85000 2.81581 0.03416* 1.2% TRI C4 (U 0.05 Zr 0.62 ,Nb 0.33 )C 0.95 6.2961 6.9031 +9.64% 4.83599 4.70004 0.12561* 2. 81% A negative value implies a loss occurred. A positive value implies a gain occur r ed Table 4 2. Uranium tri carbide pre and post HHT carbon content Sample Designation Theoretical Carbon Content (wt %) Pre HHT Carbon Content ( wt%) Post HHT Carbon C ontent ( wt%) TRI C1 9.689 10.875 11 TRI C2 9.700 0.9375 9.9425 TRI C3 9.710 9.93 9.725 TRI C4 10.32 11.775 10.3 5

PAGE 39

39 Figure 4 1. ED S of pre and post HHT TaC sample Figure 4 2. Optical image of TaC. A) Pre HHT TaC sample. B) Post HHT TaC sample

PAGE 40

40 Figure 4 3. TaC BSE image showing post HHT cracking Figure 4 4. TaC SE image showing fracture zone

PAGE 41

41 Figure 4 5 TaC SE image of pre HHT specimen Figure 4 6 TaC SE image of post HHT specimen

PAGE 42

42 Figure 4 7 TaC bulk carbon content Figure 4 8 Mid axial cross section show ing TaC carbon content as a function of depth

PAGE 43

43 Figure 4 9 TaC pre and post HHT XRD spectrum Figure 4 10 EDS of pre and post HHT of W C sample

PAGE 44

44 Figure 4 11 Optical image of WC. A) Pre HHT WC sample. B) Post HHT WC sa mple Figure 4 1 2. WC SE image of pre HHT specimen

PAGE 45

45 Figure 4 13. WC SE image of post HHT specimen Figure 4 1 4 WC bulk carbon content

PAGE 46

46 Figure 4 1 5 WC BSE image showing the transition zone where carbon depletion occurred Figure 4 1 6 Mid axial c ross section show ing WC carbon content as a function of depth

PAGE 47

47 Figure 4 1 7 Mid axial cross section of post HHT showing WC transition zone of carbon content Figure 4 1 8 WC pre and post HHT XRD spectrum

PAGE 48

48 Figure 4 1 9 EDS of pre and post HHT of Zr C sample Figure 4 20 Optical image of ZrC. A) Pre HHT ZrC sample. B) Post HHT ZrC sample.

PAGE 49

49 Figure 4 21 ZrC SE image of pre HHT specimen Figure 4 22 ZrC SE image of post HHT specimen

PAGE 50

50 Figure 4 23 ZrC BSE image of pre HHT specimen Figure 4 24 ZrC BSE image of post HHT specimen

PAGE 51

51 Figure 4 25 ZrC bulk carbon content. Figure 4 26 Mid axial cross section show ing ZrC carbon content as a function of depth.

PAGE 52

52 Figure 4 27 ZrC pre and post HHT XRD spectrum

PAGE 53

53 Figure 4 28 Optical im age of TRI C1. A) TRI C1 pre HHT. B) TRI C1 post HHT Figure 4 29 Optical image of TRI C2. A) TRI C2 pre HHT. B) TRI C2 post HHT

PAGE 54

54 Figure 4 30 Optical image of TRI C3. A) TRI C3 pre HHT. B) TRI C3 post HHT Figure 4 31 Optical image of TRI C4. A) TRI C4 pre HHT. B) TRI C4 post HHT

PAGE 55

55 Figure 4 32. TRI C1 pre and post HHT bulk carbon content Figure 4 33. TRI C2 pre and post HHT bulk carbon content

PAGE 56

56 Figure 4 34. TRI C3 pre and post HHT bulk carbon content Figure 4 35. TRI C4 pre and post HHT bulk carbon content

PAGE 57

57 Figure 4 36. TRI C1 XRD spectrum Figure 4 37. TRI C2 XRD spectrum

PAGE 58

58 Figure 4 38. TRI C3 XRD spectrum Figure 4 39. TRI C4 XRD spectrum

PAGE 59

59 Figure 4 40 TRI C1 B SE image of pre HHT specimen Figure 4 41 TRI C 1 B SE image of post HHT specimen

PAGE 60

60 Figure 4 42 TRI C2 BSE image of pre HHT specimen Figure 4 43 TRI C2 BSE image of post HHT specimen

PAGE 61

61 Figure 4 44 TRI C3 BSE image of pre HHT specimen Figure 4 45 TRI C3 SE image of post HHT specimen

PAGE 62

62 Figure 4 46 TRI C4 B SE image of pre HHT specimen Figure 4 47 TRI C4 BSE image of post HHT specimen

PAGE 63

63 Figure 4 48 TRI C1 pre and post HHT EDS Figure 4 49 TRI C2 pre and post HHT EDS.

PAGE 64

64 Figure 4 50 TRI C3 pre and post HHT EDS. Figure 4 51 TRI C4 pre and post HHT EDS

PAGE 65

65 CHAPTER 5 CONCLUSIONS AND RECO MMENDATIONS T esting the mono carbide s, WC, TaC, and ZrC in a hot hydrogen environment yi elded different results depending on the carbide composition. T he sample that had the least change in microstructu ral integrity content remained the same. No cracking, grain growth, density changes, or hydrogen embri ttlement was observed. The XRD spectrum showed the post HHT specimen to have no change in phase from the pre HH T s pecimen The TaC post HHT sample performed the second best in having minimum degradati on on the microstructural integrity There was no depletion of carbon from the sample, both bulk and cross sectional; however, the sample did experience extensive cra cking. A density decrease of 4.836% in the post HHT sample was measured against the pre HHT sample. The WC sample performed the worst in a hot hydrogen environment. The post HHT sample experienced a noticeable degree of phase change, grain boundary sep aration carbon depletion, and possible formation hydrocarbon s Portions of the post spectrum showed the formation of W 2 C. W 2 C most likely formed on the edge where carbon depletion occ urred. After a two hour test at 2775K the hydrogen was shown to deplete carbon from the WC matrix at rate of WC also experienced the greatest change in density, decreasing by 10.91%. Based on these results WC would not be a safe fuel candidate f or NTP because of the threat of fue l failure due to changes in the microstructure. Testing of the four solid solution uranium tri carbide s was also conducted at higher temperatures (approximately 2900K) for two hours, representative of NTP operating time. Analysis showed the tri carbides displayed little to no evidence of changes due to the hot hydrogen environment. The bulk carbon analysis showed that t hree of the four carbides

PAGE 66

66 displayed constant carbon content from pre to post HHT status The fourth tri carbide sample reported a los s in carbon content. This was most likely due to carbon contamination during process ing and handling of the specimen. The SEM images sho wed no regions of microstructural segregation with particular attention given to the surface since the surface is where the depletion of carbon would be the greatest. The XRD patterns verified the SEM and bulk carbon analysis by showing tha t the tri carbide specimens remain ed as unaffected solid solution carbides Based on the analysis, solid solution uranium tri carbi des appear to be compatible with hydrogen to a maxi mum tested temperature of 2926K for two hours. Hydrogen corrosion due to temperature is only o ne mechanism that can affect the rate at which hydrogen can attack a fuel matrix Future work investigating carb ides and more importantly tri carbides as NTP fuel should examine the effects of other corrosion mechanism such as fuel irradiation. Since the hot hydrogen corrosion process consist s of temperature effects coupled with radiation exposure and thermal mecha nical stresses, future test s should attempt to incorporate multiple corrosion mechanisms.

PAGE 67

67 LIST OF REFERENCES [1]. T. W. Knight Dissertation, University of Florida, Gainesville, FL 2000. [2]. S. Anghaie, T. W. carbide Fueled Reactors for Multi American Nuclear Society 2004 Winter Meeting Washington D.C., 2004 [3]. D. P. Butt, D. G. Pelaccio M. S. El Genk th Symposium of Space Nuclear Power and Propulsion Conference, Albuquerque, NM, 1994. [4]. Pewee Reactor Test Report, Lo s Alamos National Laboratory August 1969 [5]. D. R. Los Alamos National Laboratory, May, 1986. [6]. D. G. Pelaccio M. S. El Genk, D. P. Butt Carbide Power, Vol. 11 No6, November 1995. [7]. S. I. Sandlers, Chemical and Engineering Thermodynamics 3 rd Ed. John Wiley & Sons, Inc., NY, 1999. [8]. NIST JANAF Thermochemical Tables 3 RD Ed., 1985. [9]. NIST JANAF Thermochemical Tables 4 TH Ed. 1998. [10]. W. F. Seng P. A. and Carbide Formation on Iss ue 1, March 2000 [11]. J CPDS International Centre for X Ray Diffraction Data. Standard X Ray Diffraction Powder Patterns from the JCPDS Research Associateship. s.l., Newtown Square, PA 1996. [12]. Transfer St Huntsville, AL, 1981 [13]. T. W. Knight S. Processing and F abrication of M ixed U ranium/ R efractory M etal C arbide F uels with L iquid P hase S intering Elsevier, Vol. 306, 2002 [14]. B. Panda, R. R. Hickman, S. Sa ndeep Solid Solution Carbides are the Key Fuels for Future Nuc l e Huntsville, AL, 2005

PAGE 68

68 [15]. D. A. Porter, K. E. Easterling Phase Transformation in Metals and Alloys 2 nd Ed. CRC Press, Boca Raton, FL, 1992. [16]. C. R. Wang, J. M .Yang W. Hoffman, [17]. R. Zee, B. Chin, J. Cohron Auburn Univ ersity Auburn, AL, February 1993

PAGE 69

BIOGRAPHICAL SKETCH Bran don Warren Cunningham was born in 1983. He was raised in Mount Dora Florida. He graduated sum cum laude from Mount Dora High School in May 2001. Brandon attended the University of Florida and major ed in nuclear engineering. During the summer of 2006, Brandon received a fellowship with the Center of Space Nuclear Research at Idaho National Laboratory. degree from the N uclear and R adiological E ngineering D epartment a t the University of Florida in December 2006