Elevated Temperature Oxidation of Boron Modified TI-6AL-4V
Transcript of Elevated Temperature Oxidation of Boron Modified TI-6AL-4V
Wright State University Wright State University
CORE Scholar CORE Scholar
Browse all Theses and Dissertations Theses and Dissertations
2008
Elevated Temperature Oxidation of Boron Modified TI-6AL-4V Elevated Temperature Oxidation of Boron Modified TI-6AL-4V
Deborah May-Katherine Sweeney Wright State University
Follow this and additional works at: https://corescholar.libraries.wright.edu/etd_all
Part of the Engineering Science and Materials Commons
Repository Citation Repository Citation Sweeney, Deborah May-Katherine, "Elevated Temperature Oxidation of Boron Modified TI-6AL-4V" (2008). Browse all Theses and Dissertations. 818. https://corescholar.libraries.wright.edu/etd_all/818
This Thesis is brought to you for free and open access by the Theses and Dissertations at CORE Scholar. It has been accepted for inclusion in Browse all Theses and Dissertations by an authorized administrator of CORE Scholar. For more information, please contact [email protected].
ELEVATED TEMPERATURE OXIDATION OF BORON MODIFIED Ti-6Al-4V
A thesis submitted in partial fulfillment of the requirements for the degree of
Master of Science in Engineering
By
DEBORAH M. SWEENEY
B.S., University at Buffalo, SUNY, 2001
2008 Wright State University
WRIGHT STATE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
April 4, 2008_____ I HEREBY RECOMMEND THAT THE THESIS PREPARED UNDER MY SUPERVISION BY Deborah Sweeney ENTITLED Elevated Temperature Oxidation of Boron Modified Ti-6Al-4V BE ACCEPTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Master of Science in Engineering.
Raghavan Srinivasan, Ph.D. Thesis Director
George Huang, Ph.D. Department Chair
Committee on Final Examination Raghavan Srinivasan, Ph.D Allen G. Jackson, Ph.D. Sharmila Mukhopadhyay, Ph.D. Joseph F. Thomas, Jr., Ph.D. Dean, School of Graduate Studies
iii
ABSTRACT Sweeney, Deborah. M.S. Engr, Department of Mechanical and Materials Engineering, Wright State University, 2008. Elevated Temperature Oxidation Resistance of Boron Modified Titanium Alloys.
The addition of trace amounts (~0.1 wt%) of boron to titanium alloys refines the
as-cast grain size by an order of magnitude from 2000µm to 200µm. The reduced grain
size has potentially beneficial effects on the processibility of titanium alloys. Reports
also indicate that the room temperature corrosion resistance of the boron containing
alloys may be substantially greater than conventional titanium alloys. In this study, the
effects of boron addition on oxidation resistance are investigated, since conventional
titanium alloys have limited corrosion resistance in air above 650°C. The mass gain per
unit surface area was measured during elevated temperatures exposed to air to compare
the oxidation of the alpha-beta titanium alloy, Ti-6Al-4V with boron and without boron
additions. These findings are presented in conjunction with the characterization of boron
modified titanium alloys. Results include weight gain as a function of time and
temperature, activation energy calculation, microstructural characterization of the oxide
layer, measurement of oxide scale thickness and composition profiles. The experimental
results are compared to literature values for similar experiments on conventional titanium
alloys.
iv
TABLE OF CONTENTS
1 Introduction...................................................................................................................... 1
2 Background...................................................................................................................... 5
2.1 Overview Of Titanium And Alloy Use In Aerospace .............................................. 5
2.2 Boron Addition In Titanium Alloys.......................................................................... 6
2.2.1 Timeline Of Ti-B Alloys.................................................................................... 6
2.2.2 Expectations For Ti-B Alloys ............................................................................ 8
2.2.3 Processing Titanium Alloys............................................................................... 9
2.2.5 Characterization Of Titanium Alloys With Boron Addition ........................... 12
2.3 Oxidation Behavior In Titanium Alloys ................................................................. 17
2.3.1 Types Of High Temperature Corrosion ........................................................... 17
2.3.2 Thermodynamics.............................................................................................. 18
2.3.2a Oxide Scale ................................................................................................ 18
2.3.2b High Temperature Corrosion Mechanism ................................................. 20
2.3.3 Atomic Mechanism Of Oxidation.................................................................... 22
2.3.3a N-Type Cation Interstitial Oxides.............................................................. 22
2.3.3b N-Type Anion Vacancy Oxide .................................................................. 22
2.3.3c P-Type Cation Vacancy Oxides ................................................................. 23
2.3.4 Oxidation Of Conventional Titanium Alloys .................................................. 23
2.3.5 Oxidation Of Boron Containing Titanium Alloys ........................................... 34
3 Materials And Experimental Procedure......................................................................... 37
v
3.1 Materials And Factors............................................................................................. 37
3.2 Experimental Procedure.......................................................................................... 39
3.2.1 Sample Preparation .......................................................................................... 39
3.2.2 Procedure ......................................................................................................... 40
3.2.3 Microscopic Preparation .................................................................................. 43
3.2.4 Evaluation Technique ...................................................................................... 43
3.2.4a Scanning Electron Microscopy .................................................................. 43
3.2.4b Optical Microscopy.................................................................................... 44
3.3 Experimental Design............................................................................................... 45
3.4 Summary ................................................................................................................. 46
4 Results and Discussion .................................................................................................. 47
4.1 Scale Thickness....................................................................................................... 47
4.2 Scale Microstructure ............................................................................................... 50
4.3 Composition............................................................................................................ 53
4.4 Oxidation Kinetics .................................................................................................. 54
5 Summary And Conclusions ........................................................................................... 61
References......................................................................................................................... 63
vi
LIST OF FIGURES
Figure 1: Various Titanium Industries................................................................................ 2
Figure 2: Titanium-boron binary phase diagram ................................................................ 3
Figure 3: Toypta Altezza and Ti-B components ................................................................ 8
Figure 4: Ti Alloy Production Process ............................................................................. 10
Figure 5: Grain size without and with boron ................................................................... 11
Figure 6: Ti-B Phase Change Diagram ............................................................................ 11
Figure 7: Formation of β nuclei ....................................................................................... 12
Figure 8: Diffusion of boron to the grain boundaries ....................................................... 12
Figure 9: Variation of cast grain size with boron concentration ...................................... 14
Figure 10: TEM image of TiB ......................................................................................... 15
Figure 11: Schematic Ellingham Plot .............................................................................. 21
Figure 12: SEM micrograph of Ti-64 scale cross-section ............................................... 26
Figure 13: Effect of oxidation time and temperature on scale thickness ......................... 27
Figure 14: Oxide layer formation .................................................................................... 28
Figure 15: Weight gain per area of Ti-64 at select temperatures ..................................... 29
Figure 16: Oxide scale layering as a function of temperature .......................................... 30
Figure 17: Weight gain curves for oxidation of Ti-6Al-4V alloy in oxygen ................... 31
Figure 18: Curves of ( W/A)2 versus time for oxidation of Ti-6Al-4V in oxygen .......... 32
Figure 19: Arrhenius plots of parabolic rate constants .................................................... 32
Figure 20: Weight gain of Ti-64 during air oxidation ..................................................... 33
Figure 21: Oxidation Weight gain per surface area as a function of time ....................... 35
Figure 22: Arrhenius plot of parabolic rate constant ........................................................ 35
vii
Figure 23: Starting microstructures for sample materials................................................. 38
Figure 24: Sample dimensions.......................................................................................... 40
Figure 25: Experimental Apparatus .................................................................................. 41
Figure 26: Ceramic Boat sample holders.......................................................................... 42
Figure 27: SEM Micrograph............................................................................................. 44
Figure 28: Oxide Scale of PAM 5006............................................................................... 47
Figure 29: Oxide Scale of PAM 5011............................................................................... 47
Figure 30: Oxide scale of PAM 5006 for 100h at 850°C ............................................... 48
Figure 31: Oxide scale of PAM 5011 for 100h at 850°C ................................................ 48
Figure 32:Oxide Scale thickness as a function of time at 550°C...................................... 48
Figure 33:Oxide Scale thickness as a function of time at 600°C...................................... 49
Figure 34:Oxide Scale thickness as a function of time at 700°C...................................... 49
Figure 35:Oxide Scale thickness as a function of time at 850°C...................................... 50
Figure 36: Oxide pores in a sample oxidized for 50h at 850°C ....................................... 50
Figure 37: Oxide spallation on a sample oxidized for 200h at 700°C ............................ 50
Figure 38: Oxide structure, oxidized for 2 hours at 600°C............................................... 52
Figure 39: Oxide structure, oxidized for 200 hours at 600°C........................................... 52
Figure 40: Composition of PAM 5006 (without boron) at 550˚C .................................... 53
Figure 41: Composition of PAM 5011 (with boron) at 550˚C ......................................... 54
Figure 42: Weight gain/area over time at 700ºC............................................................... 55
Figure 43: Normalized weight gain as a function of time for PAM 5011 at 700ºC ......... 55
Figure 44: Log Kp vs 1/T for all materials ...................................................................... 57
Figure 45: Activation energy comparision ...................................................................... 59
viii
LIST OF TABLES
Table 1: Potential Ti usage ................................................................................................. 5
Table 2: Comparison of Coefficients of Thermal Expansion ............................................. 7
Table 3: Scale compositions of designated layers ........................................................... 26
Table 4: Materials Compositions ...................................................................................... 38
Table 5: Experimental Design .......................................................................................... 45
Table 6: Comparison of Coefficients of Thermal Expansion ........................................... 51
Table 7: Overall oxidation rate constants ......................................................................... 56
Table 8: Overall oxidation activation energy values ........................................................ 57
Table 9: Activation energy values for diffusion ............................................................... 58
Table 10: Activation energy comparison to literature values ........................................... 60
ix
LIST OF APPENDICES
Appendix A: Surface Area Measurements…………………….……………………….A-1
Appendix B: Weight Measurements and Calculated Data…….……………………….B-1
Appendix C: Graphical data………………………………….……………………...…C-1
Appendix D: Micrographs……………………………………………………………...D-1
x
Acknowledgements
Going back to college after being out for 5 years was pretty tough. I was
successful because of my own determination and stubbornness along with help from a lot
of people. I’d like to thank Dr. Bethke and Dr. Srinivasan for helping me apply, get
accepted and start classes all in the few weeks between Fall and Winter quarter. Dr
Srinivasan was instrumental throughout my entire scholastic career at Wright State. I’d
like to thank DAGSI for footing my bills and making it possible for me to go to school
full time and work part time. Thank you to my defense committee, Dr. Srinivasan, Dr.
Jackson and Dr. Mukhopadhyay for their time and expert advice. I am especially grateful
to Dr. Jackson for being a good sport and joining my committee literally at the last
minute. Greg Wilt deserves a raise for being so amazingly patient with the Cahn 1000
and all the other numerous nuisances I brought his way. Thanks for not hiding under your
desk when you heard me coming. I know you wanted to. Thanks to Mike Hall in the
Electronics Shop for all his efforts trying to get the Cahn functional even though it
proved futile. Thanks to John and Wayne in the Machine shop for helping with all my
odd requests. Balakrishna Cherukuri has been my Xenian mentor. He lent his expert
advice and highly regarded SEM talents whenever asked. He also added a cultural aspect
to my education such as explaining cricket and telling stories about all things Indian.
Most of all, thanks for being a good friend. Thanks to Satish, “Watch-out-for-that-coat-
hanger” Barney and Cutsie Buttons Baudendistel for their comic relief and for patiently
listening to me complain (and occasionally rant). Also thanks to Russell Peters for his life
lessons even if he is Canadian. “Somebody gonna get a hurt real bad.” Atlas deserves a
mention for being patient with me during the long hours he was ignored and also
accompanying me during my stress relief runs at random hours. Last, but definitely not
least, thanks to my husband Pat. He was there to support me during this whole thing and
to lend his computer nerd skills whenever needed (like when I accidentally washed my
thumb drive). Thanks for picking me up and brushing me off when I was down and
calming me when I was occasionally wound up. Everyone should be thanking you for
that last one.
1
1 Introduction
Titanium was simultaneously discovered in England and Germany in 1790.
Although titanium is abundant in the earth’s crust in the oxide form and has been found
throughout the universe, extraction is difficult. Titanium ore is found in the form of
ilmenite, titanomagnetite, rutile, anatase, and brookite but not in the metallic state.
Titanium, like aluminum, oxidizes readily when exposed to air. As a result, the extraction
of titanium requires lengthy and costly processing [1].
Titanium is known for its high strength and low density. Its density is 60% that of
steels, making it a durable and lightweight alternative. In addition, titanium is stiff, has
good toughness, good corrosion resistance, is biocompatible and works well from very
low temperatures up to as high as 593°C. Titanium can be strengthened by alloying and
deformation processes. Titanium and its alloys are used to produce aerospace components
and are useful for chemical, petrochemical, marine environments and biomedical
applications [2, 3, 4]. Figure 1 shows examples of applications including medical,
military and athletic equipment.
2
(a) (b)
(c)
Figure 1: Various Titanium Industries (a) Titanium hip implant made by Laser Engineered Net Shaping (TM) (LENS ®) (Courtesy Optomec Design Co.) (b) Litespeed bike frame (Courtesy of American Bicycle Group, Chattanooga, TN) (c) F-117 stealth: An F-117A Nighthawk warplane takes off from Holloman Air Force Base, N.M. (AP Photo/Alamogordo Daily News, J.R. Oppenheim)
The most widely used alloy is Ti-6Al-4V, commonly referred to as Ti-64, which
accounts for 60% of all titanium tonnage in the world [4]. The aerospace industry alone
uses 80% of all Ti-64 produced [5]. Because of its widespread applications in the
aerospace industry, Ti-64 is subject to a fair amount of scrutiny, as researchers
investigate ways to make the alloy stronger, stiffer and less expensive. This thesis details
one area of research, modifying Ti-64 with boron, which some believe could achieve
significant gains in these areas.
Boron modification is has several benefits. Boron and titanium form three
intermetallic compounds, TiB, TiB2 and Ti3B4 whiskers as shown in Figure 2. TiB is the
most stable of the three borides in the presence of metallic titanium. Boron is slightly
soluble (<1 atomic weight % at the eutectic) in both α and β Ti. There is a eutectic
reaction on the Ti rich side at 1813 ± 10 K a beta transus at 1157 ± 2 K [6].
3
Figure 2: Titanium-boron binary phase diagram [7]
Between 10-40 % boron, the Ti-B alloys are considered to be Ti metal-matrix
composites. Hypoeutectic, typically less than 2 weight %, are considered to be alloys [8].
Boron is an attractive additive because the coefficient of thermal expansion of TiB is
close to the value for metallic titanium. TiB also has the same density as titanium but has
a substantially higher strength and elastic modulus. Research shows boron addition
affects mechanical properties such as ductility, elongation to failure and yield strength [9,
10, 11, 12, 13]. Since there is an improvement in the mechanical properties, there is
interest in what other properties also maybe impacted.
Titanium and alloys have service temperatures up to 593°C with lower limit of
-253 °C [4]. Oxide formation and spallation make titanium and alloys unfeasible for use
at elevated temperatures without modification or coatings. This research focuses on
whether boron addition has an effect on the resistance to elevated temperature oxidation.
4
Chapter 2 covers the background information and summarizes other work on
characterization of boron modified titanium alloys. Chapter 3 details the materials used
and the experimental procedure. Chapter 4 tabulates and discusses the results obtained.
Conclusions are reached in Chapter 5.
5
2 Background
2.1 Overview of Titanium and Alloy Use in Aerospace
Titanium (Ti) is an appealing metal to designers and has been used in industry for
nearly 60 years. Nickel base and steel alloys have a density of about 1.7 times greater
than Ti alloys. Replacing the nickel base and steel alloys with Ti alloys results in
substantial weight savings that is important for aerospace applications. In addition, at
operating temperatures greater than 130°C, the use of aluminum is unfeasible, making
titanium alloys an attractive alternative [5]. Table 1 summarizes the benefits of using of
Ti alloys on aircraft.
Table 1: Potential Ti usage based on [5]
Benefits of replacement with Ti alloys
Replaces Benefits:
Steel Weight savings
Aluminum Space limitations (size/bulk factor)
Aluminum, Nickel and steel alloys Operating temperatures
Aluminum, Nickel and steel alloys Corrosion resistance
Aluminum Composite compatibility
6
The α/β alloy, Ti-6Al-4V accounts for 60% of the total titanium production today.
Approximately 80-90% of all titanium used on aircraft is Ti-6Al-4V. In pure Ti, the α
phase is stable at room temperature and the β phase is stable above 910-945±15°C (β
transus temperature) depending on oxygen content [14]. The α phase is hexagonal close
packed (HCP) and is more brittle than β, which has body centered cubic structure (BCC).
Oxygen is an α stabilizer which raises the β transus temperature. By adding alloying
elements, the stability of the two phases can be altered. Ti-64 has both aluminum, an α
stabilizer, and vanadium, a β stabilizer. At room temperature, α/β alloys contain of both
α and β phases. α/β alloys are more easily processed than other Ti alloys, can be
strengthened by solution treating and aging and are more easily welded. Together, the α
and β microstructures have higher strengths than α alone [3].
2.2 Boron addition in titanium alloys
2.2.1 Timeline of Ti-B Alloys
Although titanium was discovered in the late 1700’s, commercially pure titanium
was not developed until the early 1900’s [3]. Ferrotitanium alloys were first used in
railroad rails [15]. In the 1950’s, Armour Research Foundation and the Royal Aircraft
Establishment conducted studies on adding boron as an alloying element to improve the
stiffness of titanium alloys. The Ti-B phase diagram (Figure 2) shows that titanium
monoboride (TiB) is thermodynamically stable with both α and β phases of titanium. TiB
also has a density and CTE similar to that of metallic titanium while having stiffness and
strength substantially greater than titanium (Table 2). Therefore, TiB is ideal for
reinforcing titanium matrix composites. However, introducing TiB to titanium alloys is
7
difficult. Adding large amounts of TiB also severely decreases ductility. As a result, the
research community lost interest in boron as an alloying element for titanium [16].
Table 2: Comparison of Coefficients of Thermal Expansion based on [17, 18, 19]
Material CTE Elastic Modulus
Vickers Hardness
TiB 6.2E-6/°C 371 GPa 1800Hv
Ti 8.2E-6/°C 116 GPa 60
Boron-modified titanium appeared to be forgotten until the early 1980’s when
interest was renewed because of similarity in density and thermal expansion coefficients.
Dynamet Technologies Inc. used a blended elemental powder approach to produce a
variety of products. Primarily TiC particles were used but TiB and TiB2 were also
considered. Patents have been issued for products such as skate blades, softball bats and
automotive valves [16].
In the mid 1990’s, NASA and Boeing revisited the Ti-B research for use on the
High Speed Civil Transport with the aim to improve the structural efficiency. It was
established that large cast parts and extruded parts could be produced but the Ti-B
research in the USA was abandoned due to lack of funding. During the same time period,
Toyota was more aggressive in developing and transitioning Ti-B for automotive intake
and exhaust valves. The use of Ti-B products had lower costs and reduced spring mass.
Figure 3 shows a Toyota Altezza and the Ti-B engine valves [16].
8
Figure 3: (a) Toyota Altezza with Ti-B components (b) Ti-B Intake and exhaust valves used in the Altezza [20]
In the year 2000, the aerospace industry, in the quest for stronger and lighter
materials, again came back to Ti-B research. In a collaborative effort, FMW, Crucible
Research, AFRL/ML and others revisited the idea of modifying titanium alloys with
boron [16].
2.2.2 Expectations for Ti-B Alloys The current push in the research of boron modified titanium alloys is for
aerospace applications. However, titanium and its alloys are difficult to process and are
therefore expensive. Researchers have determined that trace amounts of boron improve
stiffness and strengthen titanium alloys by up to 20% [21]. It has been demonstrated that
introducing boron to the titanium alloy microstructure reduces grain size and potentially
improves both processibility and affordability [22].
Boron and titanium have an atomic radius ratio of 0.66. According to Hume-
Rothery Rules, there is limited mutual solubility. Therefore boron does not dissolve in the
alloy but forms TiB (TiB2, Ti3B4 at higher boron additions) [23]. No intermediate phases
form between Ti and Ti-B. At hypereutectic compositions (greater than 8 at. %) B, TiB
9
forms long single crystal needles in the Ti matrix. The TiB needles act as reinforcement
increasing stiffness and strength [21].
With the improved structural and process efficiency outlined in section 2.2.3,
these modified alloys are potentially more affordable. In cast components, improved
strength, stiffness and ductility make it possible for them to replace wrought products.
The improvement in process efficiency results in shorter forming durations. The reduced
processing time makes using superplastic forming a more practical option [24].
Overall, it is hoped that once implemented, Ti-B components will result in more
efficient, cost effective aircraft structures by improving processing and yield. In addition,
the performance will be improved through higher modulus, tensile strength, creep
properties and fatigue performance [5].
2.2.3 Processing Titanium Alloys
Titanium exists in nature as rutile which is ore composed primarily of TiO2.
Rutile is reduced, using the Kroll process, to a porous metallic titanium called “sponge”.
The sponge and alloying elements are compacted and sent through a melting process
called Vacuum Arc Remelting (VAR) to form an ingot [25]. The repetitive melting
insures compositional uniformity [1].
Figure 4 outlines the processing sequence. In the processing sequence, at the ingot stage,
the grain size is in the range of a few centimeters. The “break down forging” is necessary
to refine the grain size which increases strength, ductility and improves ease of
subsequent mechanical processing. The addition of boron results in a reduction in grain
size of the ingot to about 200 microns which makes further breakdown unnecessary [26].
10
Figure 5 compares the grain size of Ti-64 with and without boron. Figure 6 shows
a magnified section of the binary Ti-B phase diagram. The alloy, with a composition of
0.1% B, first starts out in the liquid region. As it cools, it passes the liquidus line and
forms β nuclei as shown in Figure 7. The nucleus begins to grow as the temperature
approaches the eutectic. At the eutectic temperature, β + TiB are formed. TiB is formed
at the last stage of solidification. Therefore, grain refinement in Ti-B alloys is not due to
nucleation by insoluble boride particles as in steels and aluminum alloys. The solid
interface advances and the B is rejected into the liquid in Figure 7. The solubility of
boron in titanium is negligible so TiB particles precipitate along the grain boundaries as
in Figure 8.
Figure 4: Ti Alloy Production Process based on [3]
Titanium sponge press
briquette
Manufacture of electrodes
Double Melting
Ingot
Breakdown Forging
Bloom Final
Processing
11
Without Boron
With Boron
Figure 5: Grain size without and with boron [16]
Figure 6: Ti-B Phase Change Diagram [27]
12
Figure 7: Formation of β nuclei and rejection of boron based on [28]
Figure 8: Diffusion of boron to the grain boundaries and the formation
of TiB based on[28]
2.2.5 Characterization of Titanium Alloys with Boron Addition
With the renewed interest in boron addition, a literature survey reveals various
paths taken in experimentation and characterization. Varying the amount of boron added
can result in a composite or an alloy. Studies of composites resulted in new
crystallographic information. Mechanical properties such as elastic modulus, stiffness
and microstructural stability were also investigated.
TiB containing alloys have the potential to increase stiffness by up to 20%.
Modification of the procession sequence with B implementation reduces production
costs. Improvements have been made in casting technology, superplastic forming,
diffusion bonding and welding of sheet metal and plates because of the modification [29].
Boron modified titanium alloys showed an increase in stiffness and
microstructural stability. The insolubility of B in the Ti alloy resulted in the formation of
ββ ββ
ββ ββ ββ
ββ
B B
B
B
B
B B
B
B
B
Liquid
T < Tliquidus
TiB
13
TiB whiskers crystallographically oriented with the parent lattice. The modification
significantly refined the grain size. The comparable coefficients of thermal expansion
(CTE’s) of Ti and TiB minimizes residual stresses with a thermally stable modified
metal. The fine grain equiaxed structure proved to be stable even when processed above
the beta transus in Ti-64 with 1.6% B heated to 1150°C. In addition, the Ti-TiB alloys
are able to be affordably processed using power metallurgy techniques [8]. TiB readily
formed as a primary or eutectic phase during solidification. TiB has a high elastic
modulus with a value of 371 GPa. The crystallography of the TiB/Ti interaction showed
a clean interface [17].
Motivation to explore the possibility of improved microstructures and mechanical
properties Ti-64/TiB composites spurred further research. Experimental materials were
composites containing 20-40% TiB whiskers by volume. Results showed an increase in
the elastic modulus along the extrusion axis when compared to unreinforced Ti-64. The
outcome also indicated the composite exhibited poor ductility and only moderate strength
improvement by comparison to the unmodified base alloy [30].
The grain refinement of titanium alloys modified with levels of boron below
impurity specifications for conventional titanium alloys was studied. The importance of
grain refinement on mechanical properties such as strength and ductility motivated the
research. Boron, in the amount of 0.1 weight %, was introduced to the alloy to promote
grain refinement. Experimental results showed a reduction in grain size by
approximately an order of magnitude. The refinement was attributed to the effects of the
composition change during constitutional supercooling. The change is caused by boron
rejection from the matrix to the β/liquid interface before solidification, influencing
14
nucleation and growth. Increasing the amount of boron past 0.1 weight % resulted in a
“weaker dependence of reduction in grain size” on the amount of boron (Figure 9). The
outcomes of the experimentation on composites versus low boron content alloys resulted
in the use of 0.1 weight % boron modified alloys in subsequent research [26].
Figure 9: Variation of cast grain size with boron concentration [26]
TiB typically has a whisker shape with a B27 structure with boron atoms
“zigzagging” along [010] direction. The crystallographic planes at the transverse cross-
section are always ( )100 , ( )101 and ( )110 . The TiB was hexagonal along the transverse
cross-section. TiB contains stacking faults which were observed to minimize lattice
strains at the interface between TiB and the parent lattice. Figure 10 shows TEM image
with planes, directions and stacking faults. Nucleation and growth were determined to
have an effect on the morphologies of the TiB whiskers [31]
15
Figure 10: Bright field TEM image of TiB showing the stacking faults parallel along (100) plane
The B modification raised the yield stress at room and elevated temperatures in
addition to increasing ductility at room temperature. Conclusions were reached that
borides prevent grain growth in the α phase field through the Zener Pinning Mechanism.
The borides also stabilized the lamellar structure and change the brittle to ductile
transition temperature while also increasing the creep resistance [9].
Ti-24Nb-3Al was modified with 0.05 mass % B to determine the effect on
transformation behavior and tensile properties. The results indicated superplasticity
regardless of boron addition. A higher flow stress was observed in the boron modified
alloy. The critical stress for slip, σslip, was also higher due to solid solution strengthening.
The stress to induce martensitic transformation, σsimt, increased whereas the martensitic
transformation temperature decreased. As the number of deformation cycles increased,
the σsimt decreased in both samples with and without B. However, the amount that σsimt,
16
decreased was smaller in the boron containing alloys. It was determined that boron is
effective for strengthening. It was also concluded that B suppresses the accumulation of
dislocations [32].
Different microstructures result from processing above or below the β transus
results. Mechanical properties are enhanced by controlling the microstructure through
thermomechanical processing. Titanium alloys modified with boron have proven to have
good fracture resistance, higher stiffness and strength and are currently used in non-
fracture critical applications. Results indicated a 60° increase in the beta transus due to
equilibrium and supersaturation of B in the solid solution [33].
The effects of rolling 0.1 wt. % boron modified titanium alloys into plates and
sheets were experimentally determined. The resultant microstructure and mechanical
properties were studied in rolled plates and heat treated sheets. The rolling schedules
were 75% & 92% reduction of ISM and PAM 5011 processed Ti-64 which are materials
used in the research of this thesis (specifics of processing to be discussed under materials
section). Tensile tests were performed on the rolled samples and heat treated sheets and
compared to conventional Ti-64. The results reinforced the conclusion that boron
addition decreases grain size by a factor of 10 [11, 28, 34]. On average, the modified
specimens had 10% higher yield strength, ultimate tensile strength and ductility than
conventional Ti-64 [11].
Currently, work is being done centering on titanium alloys other than Ti-64,
which contain 0.1 weight % boron. The objective is to investigate the effect of boron on
the hardenability and processability of selected beta alloys. Results to date indicate
significant grain refinement in beta titanium alloys. The current investigation is to
17
determine whether the heat treatability is altered by increased nucleation sites in the form
of increased grain boundaries and TiB/β matrix interface. Also under investigation is
whether the increased grain boundary area increases the formability of the alloy at higher
temperatures. Preliminary results indicate that despite very good grain stability at higher
temperatures, the aging processes showed no significant improvements in the mechanical
properties. Future work includes microstructural examination of the effect of boron
additions on the aging kinetics and formability studies to optimize processing conditions
for the modified alloys [34].
Collectively, research on boron containing alloys has resulted in a number of
conclusions. Large boron additions result in a loss of ductility therefore smaller amounts
such as 0.1 wt % are favored [30] The boron modification (0.1wt %) has proven to be an
effective strengthener [32]. The yield strength, ultimate tensile strength and ductility were
improved [11, 35]. Microstructural stability and stiffness, [8, 29] creep resistance [9] and
beta transus [33] also increased. Boron modifications significantly refine grain size [11,
26, 34] which decreases production costs [29]. Ti and TiB have similar CTE’s with a
clean interface [17]. There is minimal strain between the parent lattice and TiB particles
[17, 31].
2.3 Oxidation Behavior in Titanium Alloys
2.3.1 Types of High Temperature Corrosion
Corrosion experts often consider low/moderate and high temperature corrosion
to be separate phenomena because of the exponential effect of temperature on the
diffusion of reactants as shown in the Arrhenius diffusion equation. Corrosion at
18
elevated temperatures is considered in the subsequent sections. Industries affected by
high temperature corrosion include but are not limited to: heat treatment, mineral and
metallurgical processes, automotive, coal gasification, aerospace and gas turbines,
refining and petrochemical processing and nuclear. The commonly encountered primary
modes of corrosion at high temperatures are oxidation, carburization and metal dusting,
nitridation, halogen corrosion, sulfidation, ash/salt deposit corrosion, molten salt
corrosion and molten metal corrosion. Oxidation is deemed the most important high
temperature corrosion reaction because it plays a role in all industrial applications
plagued with high temperature corrosion attack [36].
Oxidation occurs in metals when they are heated in an atmosphere with air or
excess oxygen. In most cases there are sufficient oxygen activities which results in an
oxidation contribution in addition to the primary mode of corrosion. Temperature has a
direct relationship with oxidation for metals and alloys. Most metals and alloys have a
narrow operating range at high temperatures. If the ceiling temperature is exceeded, the
end result can be catastrophic oxidation [36].
2.3.2 Thermodynamics
2.3.2a Oxide Scale Most engineering metals when exposed to oxygen rapidly form a very thin oxide
layer called a scale. This scale is 3-10nm thick and has varying degrees of protection
based factors such as the metal and temperature. At high temperatures, reactants can
diffuse through the oxide and continue to thicken the oxide layer. Growth occurs by an
19
anodic or a cathodic process. In the anodic process, the metal is converted to cations
which generate electrons at the interface between the metal and the oxide. In the cathodic
process, oxygen consumes electrons at the oxide/atmosphere interface as it is converted
into anions. When an anion and a cation are brought together, the oxide scale grows [37].
Ions diffuse through the oxide via atomic defects. The Cabrera-Mott theory
outlines how the thin oxide film forms. The theory has three steps, starting with the
oxygen atoms being absorbed at the oxide/atmosphere interface. In the second step, since
the film is thin, electrons from the metal travel through by thermo-ionic emission or
tunneling. In the last step, oxygen atoms capture electrons, becoming anions. The anions
produce a strong electric field which forces the metal cations through the oxide scale
[37].
If the oxide volume and the metal volume have a large differential as growth
occurs, the film is said to be non-protective [37]. The differences produce tensile or
compressive stresses at the oxide/metal interface that may result in cracking or buckling
of the oxide scale. The Pilling-Bedworth Ratio (PBR), Equation 1, is used to predict the
sign and magnitude of the growth stress. For PBR values less than 1, a tensile stress is
metalofvolumeoxideofvolumePBR =
Equation 1
observed; for PBR values of greater than 1, a compressive stress is observed. As PBR
value deviates from a value of 1, the growth stress increases [38]. An example of this
increase in stress is the PBR in uranium which is greater than three. The resultant oxide
scale does not stick to the metal. For this reason; uranium rods in nuclear reactors must
be “canned” to prevent corrosion [37].
20
Metals that form protective oxides are controlled by Arrhenius type diffusion of
metal or oxygen through the oxide as shown in Equation 2. Diffusion (D) is dependent
on D0, the diffusion coefficient, Q, the activation energy and temperature T. The film
maintains coherency and bonds to the metal. Since D is temperature dependent, diffusion
0 exp QD DRT−⎛ ⎞= ⎜ ⎟
⎝ ⎠
Equation 2
for some metals occurs at a rapid rate which consumes the metal and makes it unsuitable
for high temperature applications. Conversely, some oxide scales withstand high
temperatures extremely well such as oxides with chromium or aluminum. Chromium and
aluminum have continuous coherent oxide films which slow diffusion of oxygen and
protect the base metal [37].
2.3.2b High Temperature Corrosion Mechanism
The driving force for metal-oxygen reactions is the free energy. If the ambient
oxygen pressure is greater than the dissociation pressure of the oxide in equilibrium with
the metal then an oxide is formed [37]. The mechanism of oxidation and other gas
corrosion modes can be outlined as shown in Equation 3 where metal M reacts with
oxygen to form a resultant oxide. Thermodynamically speaking, when the oxygen
potential in the environment is greater than the oxygen partial pressure an oxide will
2 2M O MO+ → Equation 3
form. The standard free energy of formation, ∆G°, can be used to determine the oxygen
partial pressure in equilibrium with the oxide as shown in Equation 4. The activities of
21
2
2
ln MO
M O
aG RT
a P⎛ ⎞
∆ = − ⎜ ⎟⎜ ⎟⎝ ⎠
o
Equation 4
the metal and the oxide cancel each other out because oxygen is the limiting reactant.
Solving for 2OP yields Equation 5 as seen in Talbot [37]. If the standard free energy is
known, the oxygen partial pressure in equilibrium with the oxide can also be read from an
⎟⎟⎠
⎞⎜⎜⎝
⎛ ∆−=
RTGPO
o
exp2
Equation 5
Ellingham plot. Ellingham plots are specific to selected oxides and are plotted with free
energy as a function of temperature. It is possible to determine whether or not an oxide
can form by comparing the oxygen potential in the environment with the oxygen partial
pressure in equilibrium with the oxide. Given in Figure 11, the schematic oxygen partial
pressure curve intersects the molar free energy curve for the oxidation of titanium. The
solid line is the molar free energy for the oxidation of titanium. The dashed line is the
oxygen partial pressure curve. The intersection point is equilibrium (∆G=0)--at
temperatures above the intersection temperature the metal will oxidize [39].
Figure 11: Schematic Ellingham Plot: Standard free energies of oxide as a function of temperature
22
In a comparative study of initial oxidation behavior of various titanium-aluminum
alloys, it was discovered that the oxide scale forms by the diffusion of oxygen into the
alloy bulk. The first step of the oxidation process was labeled as a dissociative
chemisorption of molecular O2. The oxygen diffused until an apparent saturation and
then became constant. It was described as “marking the end of rapid chemisorption and
initial oxide layer growth” and the “beginning of the logarithmic growth phase” [40].
2.3.3 Atomic Mechanism of Oxidation
Talbot and Talbot [37] outlined three types of oxide forming mechanisms by
defects: n-type cation interstitials, n-type anion vacancies and p-type cation vacancies.
2.3.3a N-type cation interstitial oxides
An n-type cation interstitial begins with the adsorption of an oxygen atom from
the atmosphere which is ionized to an anion, O2-. The anion is reacts with a metal cation
at the interface between the oxide and the atmosphere. The conduction band (equilibrium
gap from the valance band) supplies the electrons to ionize the oxygen atoms. The
cation is from the local interstitial population. The coordinated reaction at the interface of
the metal and the oxide is the diffusion of the cationic metal atom into the oxide along
with electrons to balance as interstitials and as a result consumes the metal. [37].
2.3.3b N-type anion vacancy oxide
Anion vacancies, n-type, expands the oxide from the metal oxide interface. A
metal atom becomes a cation and with it, a coordinating oxygen anion. The anion leaves
23
the lattice to sit on a vacancy and the free electrons join the oxide conduction band. The
reaction that takes place at the oxide/atmosphere interface adds oxygen atoms, filling in
vacancies and getting rid of free electrons [36].
2.3.3c P-type cation vacancy oxides
Metals that form p-type oxides do so by forming cation vacancies. Again, the
reaction begins with adsorption of an oxygen atom from the atmosphere which is ionized
to an anion, O2-. The anion is still coordinated with a metal cation at the interface
between the oxide and the atmosphere but the mechanism depends on the formation of
cation vacancies in the oxide. The cation sits on a lattice site and donates electrons to
ionize the oxygen, creating electron holes. The reaction is balanced at the metal/oxide
interface where metal atoms enter the oxide, donate electrons and become cations. As a
result, the cationic vacancies and holes are filled [36].
2.3.4 Oxidation of Conventional Titanium Alloys
The literature survey of relevant oxidation material covers a broad spectrum.
Researchers have speculated whether or not alloying elements have an effect on oxidation
resistance. The effect of surface preparation has also been studied. Microstructural
changes due to elevated temperature oxidation have been observed. Commercially pure
titanium, Ti-64 and other alloys have been subject oxidation at high temperatures. The
scale composition, oxygen penetration and oxide thickness in both air and oxygen
oxidation. Analysis revealed definitive temperature dependent oxidation rates and
overall activation energies. The effects of the boron addition on oxidation in previous
24
work are outlined. In the work studied, some specific experimental procedural details are
unknown such as sample dimensions and preparation.
Titanium alloy chemistry was modified to determine whether there was an effect
of composition upon the oxidation resistance. Titanium base alloys are not feasible for
use at elevated temperatures because titanium alloy surfaces become damaged by rapidly
developing scales and embrittlement due to oxygen dissolution in the metal. One
researcher insisted that alloying schemes cannot deliver a desirable level of high
temperature oxidation resistance without negatively affecting mechanical properties [41].
Other experimentation yielded contradictory results.
The effect of alloy chemistry has been examined by oxidizing Ti-A55, Ti-64 and
Ti-6Al-2Sn-4Zr-2Mo. The selected alloys were exposed at 704°C in air for 24 hours. Ti
reacted with both oxygen and nitrogen at elevated temperatures. The rate of absorption of
N was much less than that of oxygen and the Ti nitride formed was much less stable than
the Ti oxide. As a result the role of N in elevated temperature oxidation was deemed
negligible [42].
The oxidation reaction was divided into three time and temperature dependent
stages. The first stage was characterized by the formation of a dense thin layer of scale.
The movement of atoms was by anion vacancies and the rate was parabolic. In the second
stage, the scale was thick and porous and broke up which increased the oxygen exposure
to the surface. The corresponding oxidation rate was linear. During the third stage, the
porous scale sintered to the material forming a barrier between the oxygen environment
and the material. The resultant diffusion dependent oxidation rate was again parabolic.
Experimentation yielded a number of inferences about the effect of surface cleanliness,
25
environmental pressure and relative humidity. Surface cleanliness affected weight gain
by 9-17%. A sample with an acid cleaned surface gained more weight than the “dirty”
specimen. An increase from 5 torr to 760 torr resulted in a 15-32% increase in the
oxidation rate. Relative humidity also increased the rate by up to 12% with the presence
of water vapor. The grain size and variations in the microstructure had negligible effect
upon the oxidation rate. The oxygen solubility of Ti-64 at the interface between the
oxide and the metal was approximately 3 wt. %. The oxide growth on Ti-64 was greater
than the other alloys tested. This indicated that the alloying elements modified the scale
to increase the uptake of oxygen. The final conclusion reached was that at 704°C, for
24h, regardless of relative humidity, pressure or alloy, the contamination depth was 40µm
[42].
The oxidation of various alloys used in commercial titanium dental implants was
compared using XPS [43]. Foils of commercially pure Ti and Ti-64 were prepared using
different surface preparation methods: ASTM F86/B600 passivation, electropolishing in
perchloric acid based electrolyte, anodization in sulfuric acid and phosphoric acid, and
glow discharge plasma oxidation. The high resolution XPS spectra show differences in
oxide scale thickness and composition. The conclusion was reached that composition
and thickness of titanium oxide scales vary depending on the surface preparation.
The microstructural changes of Ti alloys during oxidation were studied Garbacz
et al. [44]. The experimentation involved Ti3Al, Beta-CEZ and Ti-6l-4V-0.2Fe alloys.
The experimentation consisted of annealing at 900°C for 170 hours. Examining the
results for Ti-64, SEM showed a multilayered oxide scale. EDS analysis of exposed a
V2O5 1.5 wt % band in the third sub layer. Aluminum was locally concentrated in narrow
26
layers of the scale. X ray revealed the presence of TiO2 and Al2O3 strata. Figure 12
shows the layers designated in Table 3. Table 3 outlines the compositions of the layers of
the scale. It was noted that alloys containing vanadium showed cracks and delaminations.
The cracks increased the depth of oxide penetration. The cracking and delamination were
attributed to the TiO TiO2 transition. The change in scale results in a change in
density which creates localized stress regions and decreases the overall oxidation
resistance. The resultant oxide scale was 200 µm thick. The overall weight gain was 43
mg/cm2. The oxide scale also contained a large number of defects when compared to the
other experimental materials. The final conclusion was that oxidation resistance is
heavily dependent on aluminum content.
Table 3: Scale compositions of designated layers based on [44]
Scale Composition in wt % TiO2 Al2O3 V2O5 1 98.9 - 1.1 2 98.4 - 1.6 3 95.7 2.8 1.5
Figure 12: SEM micrograph of Ti-64 scale cross-section of designated layers [44]
27
Ti-64 was oxidized in another study to determine the effect of thermal oxidation
on corrosion and corrosion wear behavior [45]. Samples were exposed to air at 600°C
for up to 60 hours. The resultant scale was composed of anatase and rutile TiO2. As
oxidation time increased, the surface roughness also increased. The oxide thickness and
penetration increased directly with time and/or temperature. Figure 13 shows the
relationship between oxidation, temperature and time, where OL is oxide layer and ODZ
is oxygen diffusion zone.
Figure 13: Effect of oxidation time and temperature on scale thickness [45]
High temperature oxidation data for Ti-64 and other alloys was lacking.
Researchers varied time and temperature in air and oxygen oxidation studies to make a
more complete compilation of oxidation data. Others explored the benefits of an
oxidized surface.
The recrystallization and beneficial thermal oxidation behavior of commercially
pure titanium (CP Ti) was examined. Research showed that the optimum temperature
range for beneficial thermal oxidation is 600-650°C. The oxygen diffusion activation
28
energy, Q, was calculated using Equation 2, where D is diffusion, Do is the diffusion
coefficient, R the gas constant and T is temperature. Q was found to be between 149 –
170 KJ/mol. Samples were oxidized for 36 and 72 hours at 600 and 650°C [46].
An oxidation model was developed aimed at predicting the lifespan of Ti alloy
components in a gas turbine engine. It was noted that the diffusion of oxygen into the
alloy was rapid at 600°C and above. Approximately 33 atomic % of oxygen was found
to diffuse into the surface metal of the substrate. The oxygen dissolved zone was labeled
the α case as shown in Figure 14 [47].
Figure 14: Oxide layer formation based on [47]
Ti-64 was oxidized in air over the range of 650-850°C. The weight change per
unit area was measured as a function of time as in Figure 15. The research established
that from 650 -700°C the oxidation rate was parabolic. From 750-800°C the rate was
parabolic-cubic. At 850°C, with less than 50 hours exposure, the rate was also parabolic-
cubic. After 50 hours, the rate became parabolic. The behavior was described in
Equation 6 where y is weight gain per area, n is the reaction index, K is the reaction
29
constant and t is time. The reaction index is the reciprocal of the slope of the log-log plot
of y versus t and varies as the rate law changes. K was calculated using Equation 7, where
Kn0 is a constant; Ea is the activation energy; R is the gas constant and T is temperature.
Air oxidation of Ti-6Al-4V
0
5
10
15
20
25
30
35
40
0 1000 2000 3000 4000 5000 6000
time (min)
wei
ght g
ain
(mg/
cm2 )
850˚C
800˚C
750˚C
650˚C700˚C
Figure 15: Weight gain per area of Ti-64 at select temperatures based on [48]
Kty n = Equation 6
Layers in the scale consisting of Al2O3 and TiO2 were observed. As the time and
temperature increased, the number of layers also increased. The oxide/environment
⎟⎠⎞
⎜⎝⎛ −
=RTE
KK ann exp
0
Equation 7
interface consisted of Al2O3 while the oxide/substrate interface consisted of TiO2 as
illustrated in Figure 16. TiO2 was found to form preferentially over Al2O3 as in evidence
at the oxide/substrate interface which is not exposed to the environment. Aluminum
30
diffuses outward as the oxygen diffuses inward to react with the titanium and thicken the
scale [48].
Figure 16: Oxide scale layering as a function of temperature based on [48]
Other oxidation research involved the oxidation behavior of Ti-64 in oxygen at
elevated temperatures. The intent was to investigate the stability of Ti-64 in a high
temperature oxygen atmosphere. The experimental durations were 2, 4, 8 and 12 hours
with at temperatures of 1050K (777°C), 1150K (877°C), 1250K (977°C) and 1340K
(1077°C). It was observed that the samples were multi-colored after oxidation. At 1250
K, for the 2 and 4 hour experiments, the scale remained intact and attached to the parent
material. For 8 and 12 hours experiments the scale was loosely bonded. At 1340 K
durations showed oxide spallation which was attributed to differential contraction
between the oxide and the substrate. Weight gain curves were plotted with weight gain
Substrate
TiO2
(a) Formation of TiO2
Substrate
Substrate
TiO2
Al2O3
TiO2
Al2O
Substrate
(b) Nucleation of Al2O3 and thickening of
TiO2 layer
(c) Formation of Al2O3 and crack
(d) Layering of Al2O3 and
crack
TiO2
Al2O
TiO2
Al2O
Substrate
crack
crack
31
per unit area (∆W/A) as a function of time (t) as per Figure 17. It was noted that as the
temperature increased, the oxidation rate increased for exposure times under 8 hours.
Figure 17: Weight gain curves for oxidation of Ti-6Al-4V alloy in oxygen [49]
Beyond 8 hours the rate reduced for temperatures 1250 K and below while for 1340 K the
rate continued to increase. 1340 K is above the beta transus for Ti 64 therefore the rate
increase is attributed to the poor oxidation resistance of β Ti-64. 1250 K was established
as a “safe” upper limit for high temperature applications of Ti-64. The parabolic
oxidation model was assumed and applied to the data. Kp, the parabolic rate constant, can
be obtained from the slope of the linear fit on Figure 18. Plotting the log of Kp verse 1/T
results in which can be fit linearly. The oxidation activation energy, Q, is the slope of the
linear fit of Figure 19 [49].
32
Figure 18: Curves of ( W/A)2 versus time for
oxidation of Ti-6Al-4V in oxygen [49] Figure 19: Arrhenius plots of parabolic rate
constants [49]
The air oxidation behavior of Ti-64 was examined over the range of 600-700°C.
The usage of Ti-64 is limited in the range of 550-600°C and above because of decreased
ductility and large amounts of oxygen dissolution. The objectives were to obtain kinetic
curves, microhardness profiles in the substrate metal and to get morphology and
composition of the oxide scale. The effects of aluminum on the composition and scale
morphology were also observed. The experimental aspects are similar to the current
research. The durations for oxidation are 50, 100, 150, 200 and 300 hours with an air
flow of 1 l/h. The oxidation was measured as mass gain per unit area over time as shown
in Figure 20 with the temperatures 600°C, 650°C and 700°C. It was determined that the
oxidation rate progressed from the time and temperature dependent diffusion controlled
rate to a linear rate. An accelerated growth in the oxide scale and a change in
morphology were observed at 650°C. It was noted that there was aluminum “pile up”
33
Weight gain of Ti-64 during air oxidation
0
1
2
3
4
5
6
0 50 100 150 200 250 300 350
time (h)
wei
ght g
ain
(mg/
cm2 )
700C650C600C
Figure 20: Weight gain of Ti-64 during air oxidation [50]
at the external surface but no continuous alumina presence through out the scale. It was
postulated that the presence of alloying elements, vanadium and aluminum reduced the
amount of oxygen dissolved in the surface of the substrate when compared to unalloyed
titanium. At 600°C and 650°C, the rate was defined as cubic. As temperature increased it
became linear as at 700°C, becoming more time dependent. The cubic to linear
transformation was attributed to acceleration in the growth rate which was accompanied
by the onset of duplex nature in the morphology at 650°C. The oxygen dissolution rate in
the substrate surface layer was determined to be parabolic. The extent of embrittlement
was measured along with oxygen diffusion coefficients. It was found that the values were
similar to the values for unalloyed Ti [50].
34
2.3.5 Oxidation of Boron Containing Titanium Alloys
Researchers oxidized various titanium alloys for select temperatures and time in
an attempt to gather more data. Experimentation involved Ti-6Al-1.2B, TiB reinforced Ti
(1:4) and TiB2 particle reinforced TiAl. In the work studied, some specific experimental
procedural details are unknown such as sample dimensions and preparation techniques.
Ti-6Al-1.2B containing in situ TiB short fibers was oxidized. The oxidation
behavior was studied over the range of 600-950°C for 50 hours. The resulting oxidation
rates were parabolic as shown in Figure 21. Using Figure 22, the overall oxidation
activation energy was found to be 250 KJ/mol. Oxygen and titanium diffusion were both
in evidence throughout the TiO2 scale. It was determined that the oxidation was
controlled by both the outward diffusion of titanium and inward diffusion of oxygen. At
750°C and up, B2O3 gas forms and creates pores. The porous layer becomes cracked at
850°C and above due to an increased number of pores at the Ti/TiB interface and in the
matrix. It was concluded that in situ TiB decreases the oxidation resistance due to the
formation and evaporation of B2O3 [51].
35
Weight gain per surface area vs. time
0
5
10
15
20
25
0 500 1000 1500 2000 2500 3000 3500
time (min)
wei
ght g
ain/
area
(mg/
cm2 )
950˚C
900˚C
800˚C750˚C
700˚C
Figure 21: Oxidation Weight gain per surface area as a function of time [51]
Q calculation
-6
-4
-2
0
2
4
6
8.0E-04 8.5E-04 9.0E-04 9.5E-04 1.0E-03 1.1E-03 1.1E-03 1.2E-03 1.2E-03
1/T (K-1)
ln K
p (m
g2 /cm
4 min
)
Figure 22: Arrhenius plot of parabolic rate constant for the oxidation of Ti-6Al-1.2B over 600-950°C [51]
36
The early stage oxidation of TiB reinforced Ti composites at high temperatures
was investigated [52]. The experimental temperatures were 550, 600 and 650°C. The
environmental condition was purified oxygen at atmospheric pressure. It was found that
at the initial onset of oxidation, ~0.8 ks (13.33 minutes), the oxidation rate was linear.
Later oxidation rates followed a parabolic curve. Experimental results indicated that as
the oxidation time increased, the oxidation rate decreased. The resultant oxide scale was
composed of rutile. It was concluded that the presence of the TiB reinforcement
decreased the oxidation rate at the selected temperatures. The decrease was attributed to
the interface cohesion between the TiB reinforcement and the matrix. Also, the clean
microstructure at the interface acts as a barrier to solid state diffusion.
The oxidation of TiB2 particle reinforced TiAl intermetallics was also researched
[53]. The experimental materials contained varying amounts of TiB2 dispersed particles,
0, 3, 5 and 10 wt. %. The selected temperatures ranged from 800 to 1000°C.
Experimental results indicated diffusion of Ti outward and diffusion of O inward
resulting in a triple layered scale of TiO2, Al2O3, and mixed TiO2 + Al2O3. No B2O3 was
observed due to evaporation. It was concluded that the addition of TiB2 particles
increased the oxidation resistance of TiAl alloy. The resistance also increased as the
amount of TiB2 added increased. The increase was attributed to the TiB2 enhancement of
alumina forming tendencies. The TiB2 also provides additional nucleation sites resulting
in the formation of a thin, dense oxide scale.
37
3 Materials and Experimental Procedure
Chapter 3 is divided into the following two sections: Materials and Factors,
Experimental Procedure and Experimental Design. The Materials and Factors section
identifies the experimental and control materials and outlines the manufacturing
processes and compositions of each. It also indicates other factors that affect the
experimental outcome. The Experimental Procedure section details sample preparation,
the oxidation experiment and microstructural characterization. The Experimental Design
outlines the experimental conditions each material was subject to.
3.1 Materials and Factors
The materials used are outlined in Table 4. The experimental materials consisted
of Ti-64 produced two ways, Induction Skull Melting (ISM) and Plasma Arc Melting
(PAM). The ISM Ti-64 is a boron modified alloy also with 0.1 weight % boron. ISM is
produced by Flowserve and is subject to hot isostatic pressing (HIPing). The
experimental ISM is as cast in ingot form. The ingot was sliced into discs for the
experimental preparation. PAM 5011 is another variation of the Ti-64 alloy with an
addition of 0.1 weight % boron. The PAM 5011 alloy is produced by RTI International
Metals. One of the experimental control materials is mill product Ti-64 which is
produced by TIMET using standard titanium alloy processing techniques. The material
for the experiment was cut into a disc for further preparation. The other control is PAM
5006, which is not modified by boron but is also produced by RTI. Both PAM materials
38
were cut from a plate rolled to 75% reduction. Figure 23 contains starting microstructure
micrographs for each material.
Table 4: Materials Compositions
Material Manufacturer Composition
ISM Flowserve Ti-6Al-4V-0.1B
PAM 5011 RTI Ti-6Al-4V-0.1B-0.18O PAM 5006 RTI Ti-6Al-4V-0.08
Ti-6Al-4V mill product TIMET 6.45Al+4.02V+0.14Fe+0.18O
PAM 5006 PAM 5011
ISM Ti-64 Mill Product
Figure 23: Starting microstructures for sample materials [11]
39
Factors of the experiment were temperature and time. The temperatures chosen
were 550, 600, 700 and 850ºC. The lowest temperature, 550ºC was chosen because it is
approximately the upper limit for feasible operating temperatures for Ti-64. The highest
temperature, 850ºC, was chosen because the alloy may be exposed to this temperature for
short durations in service and also during processing. The temperature was monitored
using a digital output from a K type thermocouple centered in the heat zone of the
furnace. The oxidation durations of the experimental runs were 2, 10, 20, 50, 100 and 200
hours.
3.2 Experimental Procedure
3.2.1 Sample Preparation
The sample surfaces were prepared with 240 grit paper to remove any tool marks,
cutting residue and other impurities. The samples were then ultrasonically cleaned, first
in water to remove grinding debris then with acetone to remove any oils from handling.
Each sample was dried thoroughly then weighed and dimensioned.
All material dimensions were measured in the same manner. The ISM, PAM
5006 and PAM 5011 slabs were milled into a 4mm thick sheet. Using electrical
discharge machining (EDM), the slabs were cut into 10mm strips. The Ti-64 control had
an original thickness of 6.25mm and was cut into 10mm strips using a diamond blade.
The strips were then cut into approximately 3.5mm sections with a diamond blade
making the final sample dimensions approximately 4 mm in length, 4 mm in width and
10 mm in height as shown in Figure 24. The sample dimensions were measured using
Mitutoyo digital calipers. Final dimensions were taken for later use in surface area
40
Figure 24: Sample dimensions
calculations. Equation 8 was used to calculate the surface area for each sample. The
surface area was calculated assuming the faces of the rectangle are trapezoidal to account
for small variances in cutting each face. The weight of each sample after cleaning but
22112121
22
22 wtwtwwLttL ++⎟
⎠⎞
⎜⎝⎛ +
+⎟⎠⎞
⎜⎝⎛ +
Equation 8
before oxidation was measured using a Mettler AE100 balance with 0.1mg sensitivity.
After oxidation, the sample was cooled then weighed again. The difference between the
pre and post oxidation weights is the weight gain. The weight gain per unit area was
found by dividing the gain by the surface area. The resulting quantity is referred to as the
normalized weight gain.
3.2.2 Procedure
The oxidation process was carried out in a horizontal furnace with a controlled
environment at atmospheric pressure. Figure 25 shows the experimental set up. The
samples were contained in ceramic boat sample holders pictured in Figure 26. A dry air
atmosphere was maintained by using a slow, constant flow rate through the system. The
41
air atmosphere was passed through two Drierite hygroscopic packed beds of anhydrous
calcium sulfate before entering the system. The air flow was approximately 0.4 l/min.
The constant flow rate was controlled using a regulator and flow meter. The oxidation
durations were 2, 10, 20, 50, 100 and 200 hours run at 550, 600, 700 and 850ºC. After
each experimental period, the samples were removed and allowed to air cool on a
ceramic surface. Upon reaching ambient temperature, the samples were weighed again to
determine the weight gain.
Figure 25: Experimental Apparatus
42
Figure 26: Ceramic Boat sample holders
43
3.2.3 Microscopic Preparation
Samples were mounted using a Buehler Simplimet mounting press in Buehler
Konductomet powder according to manufacturer’s directions. Using a Buehler 10 sample
mount, the following grinding/polishing procedure was run on each sample. Each cycle
included 3 sets of 2 minute grinding segments with a constant water stream. The first
cycle used 240 grit paper turning on a complementary revolving disc at 160 RPM with no
applied pressure. The same settings were used for 320, 400 and 600 grit paper. The finer
polishing segments used Texmet 2500 polishing cloth with liquid soap and diamond paste
applied. The polishing sequence include a cycle of 9µm, 6 µm, and 3 µm diamond paste
for 5 minutes with 1.8lbf (8.0N) pressure gradually increasing 3lbf (13.3N) pressure for
the final cycle with 1µm diamond paste for 8 minutes. The samples were clean between
cycles with liquid dish soap. After the polishing sequence, the final polish was done in a
Vibromet 1 with 0.05 micron colloidal silica solution for approximately 24 hours. The
samples were cleaned with soap then rinsed with methanol and air dried.
3.2.4 Evaluation Technique
3.2.4a Scanning Electron Microscopy
Scanning Electron Microscopy (SEM) allowed for examination of the order of µm
of the scale and base metal. The oxide scale thickness was measured using micrographs
taken with Jeol SEM Model JSM-7401s Field Emission Scanning Electron Microscope.
The image analysis software, Scandium [54], which was supplied with the microscope
44
was used to calculate the average oxide layer thickness, which was then plotted as a
function of time for each material. Energy dispersive x-ray analysis (EDAX) was used to
determine the elemental composition of the oxide layer and the base metal at different
depths from the specimen surface.
Qualitatively, the scale and metal was inspected for voids and other large
abnormalities. Micrographs were obtained for further study of the resultant oxide layer
structure. The number of definitive layers formed and the thickness were correlated to
the level of oxidation resistance. The micrographs allow a visual inspection of the
adhesion of the layers and also the morphology shown in Figure 27.
Figure 27: SEM Micrograph
3.2.4b Optical Microscopy
Optical microscopy allowed for examination of the order of µm up to mm of the
scale and base metal. The layers were quantitatively analyzed. The thickness of each
layer was also measured. Qualitatively, the scale and metal was inspected for voids and
other large abnormalities. Micrographs were obtained for further study of the resultant
Oxide
Base Metal
45
oxide layer structure. The number of definitive layers formed and their thickness were
correlated to the level of oxidation resistance. Microhardness tests were conducted on a
limited number of samples to track hardness changes as a function depth.
3.3 Experimental Design
The design of the experiment included a full factorial representation with some
repetition. Table 5 outlined how all materials were subject to the same experimental
conditions differentiating between the runs with repetition. The “O” denotes full factorial
without repetition while the “X” denotes full factorial with repetition. Repetition was not
feasible in some cases due to a limited supply of the experimental material.
Table 5: Experimental Design
Oxidation Time (h)
Material Temperature 2 10 20 50 100 200
550 O O O O O O 600 O O O O O O 700 O O O O O O
Ti-64
850 O O O O O O 550 X X X X X X 600 X X X X X X 700 X X X X X X
PAM 5006
850 O O O O O O 550 X X X X X X 600 X X X X X X 700 X X X X X X
ISM
850 O O O O O O 550 X X X X X X 600 X X X X X X 700 X X X X X X
PAM 5011
850 O O O O O O
X: repetition O: no repetition
46
3.4 Summary
The metrics, surface area and weight gain, were used to calculate a normalized
weight gain. The calculated weight gain plotted against the factors time and temperature,
allowed the graphical calculation of Kp (the over all oxidation rate constant) and
ultimately Q, the overall oxidation activation energy. The metric scale thickness allowed
for a correlation of rate of oxide growth over time. The micro hardness metric allowed
for measurements of oxygen penetration and alpha case depth. The metric compositions
allowed a hypothesis to be formed about the atomic behavior during oxidation which
ultimately resulted in different layer compositions. The material factors allowed a
comparison to determine whether or not the addition of born affected the oxidation
resistance of Ti-64 using experimental and control materials.
47
4 Results and Discussion
Chapter 4 includes the results of experimentation and a discussion of the results.
The sections detail the results of overall oxidation activation energy calculations,
measurement of the oxide scale, analysis of the scale, and compositions profiles. The
results of each are discussed and compared with relevant literature.
4.1 Scale Thickness
Using the Scanning Electron Microscope (SEM), the change in oxide layer
thickness was measured as a function of time. Figure 28 and Figure 29 are SEM
micrographs of PAM 5006 and 5001, respectively, oxidized at 700ºC. The SEM was used
to measure the oxide thicknesses for all materials and times for 550, 600 and
Figure 28: Oxide Scale of PAM 5006 (without boron) for 200h at 700°C
Figure 29: Oxide Scale of PAM 5011
(with boron) for 200h at 700°C
TiB
48
700°C. An optical microscope was used to measure scales oxidized at 850°C for all
materials. Figure 30 and Figure 31 are optical micrographs for PAM 5006 and PAM
5011 oxidized for 100 hours at 850°. Figure 32 through Figure 35 contain plots of the
oxide scale growth as a function of time. The plotted data follows a parabolic curve
which becomes steeper with increasing temperature.
Figure 30: Oxide scale of PAM 5006 for 100h at 850°C
Figure 31: Oxide scale of PAM 5011 for 100h at 850°C
Oxide growth vs time @ 550˚C
0
0.5
1
1.5
2
2.5
0 50 100 150 200 250
time (h)
thic
knes
s (m
)
50065011ISMTi-64
Figure 32:Oxide Scale thickness as a function of time at 550°C
Mount
Oxide
Gap
Metal
Mount
Oxide
Gap
Metal
49
Oxide growth vs time @ 600˚C
0
1
2
3
4
5
6
0 50 100 150 200 250
time (h)
thic
knes
s ( µ
m)
50065011ISMTi-64
Figure 33:Oxide Scale thickness as a function of time at 600°C
Oxide scale thickness vs time @ 700˚C
0
5
10
15
20
25
30
35
40
45
0 50 100 150 200 250
time (h)
scal
e th
ickn
ess
( µm
)
5006
5011
ISM
Ti-64
Figure 34:Oxide Scale thickness as a function of time at 700°C
50
Oxide scale thickness vs time @ 850˚C
0.00
100.00
200.00
300.00
400.00
500.00
600.00
0 50 100 150 200 250
time (h)
thic
knes
s (m
m)
50065011ISMTi-64
Figure 35:Oxide Scale thickness as a function of time at 850°C
4.2 Scale Microstructure
As temperature and time increased, the scale also became more porous, as in
Figure 36, and began to separate from the base metal, forming a gap at the oxide/metal
interface as shown in Figure 37. At increased temperature and time, the scale became
100µm100µm
Figure 36: Oxide pores in a sample oxidized for 50h at 850°C
(Optical Mag: 10x)
Figure 37: Oxide spallation on a sample oxidized for 200h at 700°C
(SEM Mag: 250x)
51
more brittle and spalled off with the slightest disturbance. As a result, some scale
thickness measurements could not be made. The delamination and cracking observed at
increased temperature and time were also observed by Garbacz et al. [44] and Mungole et
al. [49]. Mungole et al. attribute the spallation to differential thermal contraction between
oxide and base metal. According to Guleryuz et al. [45], the scale is primarily composed
of anatase and rutile TiO2. The Coefficient of Thermal Expansion (CTE) for rutile,
anatase and Ti-64 are compared in Table 6. There is a small differential between the
values, however, it is unlikely that thermal expansion differences are entirely responsible
for the spallation.
Table 6: Comparison of Coefficients of Thermal Expansion based on [55]
CTE linear 20˚C µm/m-˚C
rutile 7.14
anatase 10.2
Ti-6Al-4V 8.6
A more plausible cause for the spallation is large growth stress which is often
described using the Pilling-Bedworth Ratio (PBR). It can be noted that although there is
not a direct relationship between PBR and stress, growth stresses can be predicted from
it. A large differential from 1.0 indicates large growth stresses [38]. The calculated PBR
value for the TiO2/Ti relationship is 1.80. The differential of 0.8 indicates the growth
stresses present are not significant.
The spallation is most likely due to a combination of the previously mentioned
factors: scale porosity, PBR value, and the small differences in CTE’s. TiO2 has a high
52
Vickers hardness and is very brittle. As a result, the oxide scale cracks and buckles when
subject to stresses.
Figure 38 and Figure 39 show the growth of the oxide for samples exposed to
600ºC for 2 and 200 hours. The particles in the micrograph form the oxide scale. As time
and temperature increase, the particle size increases. Zhang et al. [51] attribute these
pores to the formation of B2O3 gas which in turn evaporates and leaves a void. There is
no evidence of boron in the oxide scale so the formation of B2O3 gas cannot be
discounted. However, experimental materials contain only 0.1 wt. % B. Since the pores
1µm1µm
1µm1µm
Figure 38: Oxide structure, oxidized for 2 hours at 600°C
Figure 39: Oxide structure, oxidized for 200 hours at 600°C
observed can be viewed under low magnification, the trace amount of B2O3 gas is not a
viable cause of the pore formation. A more likely cause is the increased particle size
which leaves larger spaces between particles as size increases with time and temperature.
Zhang [51] also concluded that the pore formation lowered the oxidation resistance. As a
result, it was also concluded that boron addition reduces oxidation resistance. Since all
samples, both with and without boron exhibited pore formation, it was determined that
boron does not cause a reduction in oxidation resistance.
53
4.3 Composition
Using the SEM EDAX, composition profiles were plotted for each material.
Figure 40 and Figure 41 show weight % of each element present as a function of distance
from the edge of the sample (mount/oxide interface) for PAM 5006 and PAM 5011
respectively, oxidized at 550˚C. Profiles for all sample materials and the selected
temperatures maybe viewed in Appendix B. The sharp decrease in oxygen concentration,
accompanied by the sharp increase in titanium concentration took place in most cases
within the first 100µm from the edge. EDAX showed constant compositions of
aluminum and vanadium across the sample body, indicating the presence of small
amounts of Al2O3 and V2O5. The profile of oxygen content decreases toward values of
~2 wt %.
5006 composition 550°C
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14
distance (µm)
wt %
O%Ti%Al%V%
Oxide Metal
Figure 40: Composition of PAM 5006 (without boron) at 550˚C
54
5011 composition 550°C
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14
distance (µm)
wt %
O%Ti%Al%V%
Oxide Metal
Figure 41: Composition of PAM 5011 (with boron) at 550˚C
4.4 Oxidation Kinetics The normalized weight gain for each sample was obtained by dividing the
measured weight gain by the total surface area. Figure 42 shows the normalized weight
gain per unit surface area as a function of time for all sample materials at 700ºC. In order
to determine the rate constant for parabolic growth, the normalized weight gain squared,
(∆W/A) 2, was plotted as a function of time. Figure 43 shows (∆W/A) 2 verse time for
PAM 5011 at 700ºC. Using the calculated slopes, Kp was determined at 700ºC for each
material. (∆W/A)2 verse time plots were made for PAM 5006, ISM and Ti-64 at 550, 600
and 850ºC which can be viewed in Appendix B. The rate constants for all materials at
each temperature are tabulated in Table 7.
55
Comparison W/A vs time T=700°C
0
0.00001
0.00002
0.00003
0.00004
0.00005
0.00006
0.00007
0.00008
0 20 40 60 80 100 120 140 160 180 200
time (h)
W/A
(g/m
m2 )
5006ISM5011Ti-64
Figure 42: Weight gain/area over time at 700ºC
Kp (5011) @ 700oC
0.0E+00
5.0E-10
1.0E-09
1.5E-09
2.0E-09
2.5E-09
3.0E-09
3.5E-09
0 50 100 150 200 250
time (h)
(W/A
)2 (g
/mm
2 )2
Figure 43: Normalized weight gain as a function of time for PAM 5011 at 700ºC
56
Table 7: Overall oxidation rate constants
Kp values (g2/mm4h)
550ºC 600ºC 700ºC 850ºC PAM 5006 3.17E-14 1.86E-13 1.80E-11 5.75E-09
ISM 4.20E-14 1.44E-13 1.27E-11 4.19E-09 PAM 5011 3.35E-14 2.23E-13 1.44E-11 4.22E-09
Ti-64 7.08E-14 4.66E-13 2.04E-11 2.08E-09
As per the work of Du et al. [48] in Equation 7, the calculated Kp values were
collected for each material and plotted with log Kp on the y axis and T -1on the x axis.
The slope of the plot is –Q/R which can be solved for Q, the overall oxidation activation
energy. The overall oxidation energy includes the contributions from oxygen, titanium,
⎟⎠
⎞⎜⎝
⎛ −=
RTE
KK app exp
0
Equation 7
aluminum and vanadium diffusion through the base metal and oxide scale. In Figure 44,
log Kp is plotted as a function of T -1. The calculated Q values are summarized in Table 8.
There is no significant difference between the alloy Q values.
57
Kp vs 1/T
y = -1.2839x + 2.4431 (Ti-64)
y = -1.554x + 5.285 (5011)
y = -1.6136x + 5.9879 (5006)
y = -1.553x + 5.2726 (ISM)
-14
-13
-12
-11
-10
-9
-88.5 9 9.5 10 10.5 11 11.5 12 12.5 13
104/T (K-1)
log
Kp
(g2 /m
m4 s)
5006ISM5011Ti-64
Figure 44: Log Kp vs 1/T for all materials
Table 8: Overall oxidation activation energy values
Material Activation Energy
Ti-64 115.2 KJ/mol
ISM 131.3 KJ/mol
PAM 5011 131.7 KJ/mol
PAM 5006 136.7 KJ/mol
The overall oxidation activation energy indicates how easily the base metal will
oxidize. Higher activation energy indicates a higher resistance to oxidation. Qin et al.
[52] and Lee et al. [53] oxidized Ti alloys with a high boron content (composites). Both
researchers concluded that the presence of TiB increased the oxidation resistance.
However, at the level of 0.1 wt % boron, no effect on the oxidation resistance due to the
boron addition was observed.
58
Table 9 lists various activation energy values of the most prevalent reaction in the
oxidation of Ti-64. The tabulated values are in some cases twice the value of the
experimental values. One possible explanation for the discrepancy is the tabulated values
are individual activation energies as opposed to overall activation energies which take
into account all diffusion rates and is controlled by the slowest step. The experimental
activation energy values are overall and are the result of the combined effect of the
tabulated values in addition to Al, O, B, Ti and V diffusion in Al2O3, TiO2, V2O5, B2O3,
TiB and/or Ti-6Al-4V.
Mungole, Frangini, Zhang and Du et al. [49, 50, 51, 48] worked on similar
oxidation studies and will be used as comparisons. Each author used select elevated
temperatures greater than or equal to 600˚C. The normalized weight gain for each data
series is along the same order of magnitude although plotted for assorted time periods.
Figure 45 contains PAM 5006 and PAM 5011 data for log Kp as a function of T-1
Table 9: Activation energy values for diffusion based on [56]
Alloy (wt.%) Activation energy (kJ/mol)
Pure Ti self-diffusion 239
Ti diffusion in TiO2 257
O fast diffusion path in Al2O3 241
plotted against Mungole, Frangini, Zhang and Du et al. [49, 50, 51, 48]. The resultant
trends are of the same order of magnitude. The slope of Figure 45 is used to calculate
the overall activation energy, Q as shown in Equation 7. Table 10 contains calculated
values for the experimental materials, PAM 5006 and PAM 5011, Mungole, Frangini,
59
Zhang and Du et al. [49, 50, 51, 48]. The experimental values fall in the range of the
literature values. Discrepancies are due to differing alloy compositions [42], preparatory
techniques [42, 43] and environmental conditions such as atmosphere and temperature.
Comparison to Literature Q plots
y = -1.0025x - 0.432 (Mungole)
y = -1.2639x + 2.0919 (Du)
y = -1.589x + 5.6506 (5011)
y = -1.6414x + 6.2453 (5006)
y = -3.3439x + 14.792 (Zhang)
y = -0.7649x - 1.7708 (Frangini)
-25
-20
-15
-10
-5
06 7 8 9 10 11 12 13
104/T (K -1)
log
Kp (
kg2 /m
m4 h)
DuMungole50065011ZhangFrangini
Figure 45: Activation energy comparision: Frangini, Du, and Zhang [48, 50, 51], PAM 5006 and PAM 5011 at 700˚C
60
Table 10: Activation energy comparison to literature values of Mungole, Frangini, Zhang and Du et al. [49, 50, 51, 48]
Author Activation energy (KJ/mol)
Frangini 63.6
Mungole 83.3
Du 105.1
PAM 5006 136.5
PAM 5011 132.1
Zhang 250.0
61
5 Summary and Conclusions
The objective was to determine whether the addition of 0.1 wt. % boron had an
effect on the oxidation resistance of selected Ti-64 alloys through experimentation and
analysis. The weight gain over time at selected temperatures was measured and used to
calculate overall oxidation activation energy for each sample material. The calculated
value was compared among the experimental materials and with other research values.
As further analysis, the scale thickness for each material at the selected temperatures and
times was measured. The scale composition was also plotted in the form of a
composition profile for comparison.
At individual temperatures, materials could be ranked from lowest gain to highest
gain. The Ti-64 mill product most consistently had the largest gain which could be
speculated to be due to the iron content. The highest gains overall translated to the lowest
activation energy. The other materials were less consistent in the relative gain which
translated to similar activation energies. Though the activation energy was lower for Ti-
64, it was not significantly so. As a result it was determined that the boron addition did
not have an effect on the oxidation resistance.
Analysis of scale thickness showed a parabolic gain with temperature. The
differential in weight gain translated to the same gain as the thickness. The scale itself
was qualitatively analyzed. The pores, cracks and delamination were not material
specific but were dependent on time and temperature.
62
The composition profiles were plotted with weight % as a function of distance
from the sample edge. The material specific plots were similar in nature and did not
suggest any difference in oxygen diffusion between them. As expected, the lower time
and temperature samples experienced a sharp decrease in oxygen content accompanied
by a sharp increase in titanium content sooner than higher temperatures and times.
In conclusion, this study shows that the addition of 0.1wt% boron to Ti-64 does
not affect the oxidation resistance of the alloy over the temperature range 550º to 850ºC.
Therefore, the processing benefits from the early stage grain refinement resulting from
trace boron additions are attainable without any detrimental oxidation for the most
commonly used titanium alloy Ti-6Al-4V. Further characterization of boron modified
alloys that may be done involve fatigue, creep and corrosion.
63
References
[1] R. Boyer, G. Welsch, E.W. Collins, Eds., Materials Properties Handbook:
Titanium Alloys, Materials Park, OH: ASM International [2] M.J. Donachie, Jr., "Titanium," Metals Handbook Desk Edition, Ed. H.E. Boyer
and T.L. Gall, Metals Park, OH: American Society for Metals, 1985, pp. 9.1-9.12 [3] M.J. Donachie, Jr., "Introduction to Titanium and Titanium Alloys," Titanium and
Titanium Alloys, Ed. M.J. Donachie, Jr., Metals Park, OH: American Society for Metals, 1982, pp. 3-9.
[4] M.J. Donachie, Jr., "Titanium and its alloys," Titanium and Titanium Alloys, Ed.
M.J. Donachie, Jr., Metals Park, OH: American Society for Metals, 1982, p 15. [5 ] Boyer R.R, "An overview on the use of titanium in the aerospace industry,"
Materials Science and Engineering: A,vol. 213, no. 1, pp. 103-114, 1996. [6] J.L. Murray, P.K. Liao and K.E. Spear, Binary Alloy Phase Diagrams , Ed. T.B.
Massalski, Metals Park, OH: ASM, 1986, pp. 387-392. [7] H. Baker, Ed., ASM Handbook: Alloy Phase Diagram. Metals Park, OH,
USA: ASM International, 1992, vol. 3. [8] K.S. Ravi Chandran and D.B. Miracle, "Titanium-Boron Alloys and Composites:
Processing, Properties and Applications," Journal of the Minerals, Metals and Materials Society, vol. 56, no. 5, pp. 32-33, May 2004.
[9] Z.X. Li and C.C. Cao, "Effects of minor boron addition of phase transformation
and properties of Ti-47.5Al-2Cr-2Nb alloy," Intermetallics, vol. 13, no. 3-4, pp. Pages 251-256, March-April 2005.
[10] S. Tamirisakandala, R.B.Bhat, J.S. Tiley, D.B. Miracle, "Microstructural
Refinement of Titanium Alloys via Boron Addition," Journal of Materials Engineering and Performance, vol. 14, no. 6, pp. 741-746, December 2005.
[11] M.D. Bennett. Processing and property improvement in rolled plates and sheets of
Ti-6Al-4V + 0.1 WT% B. Master’s Thesis, Wright State University, 2006 [12] T. M.Godfrey, P. S. Goodwin, C. M. Ward-Close, "Titanium Particulate Metal
Matrix Composites - Reinforcement, Production Methods, and Mechanical Properties," Advanced Engineering Materials, vol. 2, no. 3, pp. 85-91, 2000.
64
[13] D.B. Miracle, "Titanium-Boron Alloys," Metallic Materials with High Structural
Efficiency, Ed. O.N. Senkov, D.B. Miracle and S.A. Firstov, Springer, 2004, pp. 10-13.
[14] M.J. Donachie, Titanium: A Technical Guide, 2nd ed., Materials Park, OH: ASM
International, 2000. [15] W.H. Sellew, Railway Maintenance Engineering: With Notes on Construction,
New York: D. Van Nostrand Company, 1915. [16] D. Miracle, S. Tamirisa, R. Bhat, R. Srinivasan and K. Schwendiman, Ti-B
ALLOY TECHNOLOGY: An Introduction and Overview. Ti-B Alloy Workshop Dayton, OH. October 2005
[17] K.B. Panda and K.S. Ravi Chandran, "Synthesis of Ductile Titanium-Titanium
Boride (Ti-TiB) Composites with a Beta-Titanium Matrix: The Nature of the TiB Formation and Composite Properties," Metallurgical and Materials Transactions A, vol. 34A, no. 6, pp. 1371-1385, June 2003.
[18] K.B. Panda and K.S. Ravi Chandran, " First principles determination of elastic
constants and chemical bonding of titanium boride (TiB) on the basis of density functional theory," Acta Materialia, vol. 54, no. 6, pp. 1641-1657, April 2006.
[19] Automation Creations Inc., MatWeb Material Property Data, [Online Database],
Copyright: 1996-2008, Available at: http://www.matweb.com/search /DataSheet.aspx?MatID=115.
[20] T. Saito, "The automotive application of discontinuously reinforced TiB-Ti
composites," Journal of the Minerals, Metals and Materials Society, vol. 56, no. 5, pp. 33-36, May 2004.
[21] O. Ivasishin, R.V. Teliovych, V.G. Ivanchenko, S. Tamirisakandala, and D.B.
Miracle. Processing, microstructure, and tensile properties of the Ti-6Al-4V-1.55B eutectic alloy, AFRL-ML-WP-TP-2007-438.Materials and Manufacturing Directorate Air Force Research Laboratory, Wright-Patterson Air Force Base, OH. February 2007
[22] K.S. Ravi Chandran and D.B. Miracle, "Titanium-Boron Alloys and Composites:
Processing, Properties and Applications," Journal of the Minerals, Metals and Materials Society, vol. 56, no. 5, pp. 32-33, May 2004.
[23] M.E. Hyman, C. McCullough, J.J Valencia, C.G. Levi and R. Mehrabian, "
Microstructure Evolution in TiAl Alloys with B additions: Conventional Solidification," Metallurgical Transactions A, vol. 20A, no. 9, pp. 1847-1859, Sept 1989.
65
[24] D.B Miracle, S. Tamirisakandala, R.B. Bhat and J.S. Tiley. Titanium Alloy Microstructural Refinement Method And High Temperature, High Strain Rate Superplastic Forming Of Titanium Alloys. World Intellectual Property Organization. WO/2005/060631. 7Jul07.
[25] G. Lütjering and J.C. Williams, Titanium, First ed., Springer, 2003. [26] S. Tamirisakandala, R.B. Bhat, J.S. Tiley and D.B. Miracle, "Grain refinement of
cast titanium alloys via trace boron addition," Scripta Materialia , vol. 53, no. 12, pp. 1421-1426, December 2005.
[27] J.L. Murray, Phase Diagrams of Binary Titanium Alloys, First ed., Metals Park,
Ohio: ASM International, 1987. [28] S. Tamarisakandala, R.B. Bhat, D. McEldowney, J.S. Tiley and D.B. Miracle.
“Microstructural Refinement of Titanium Alloys via Boron Addition.” Materials and Manufacturing Directorate, Air Force Research Laboratory. AFRL/WS 05-1287. Oct2005
[29] F. Doerner and J. Simpson, "Titanium alloy developments: Trends in titanium
structures for aircraft of the future," Advanced Materials and Processes, vol. 164, no. 5, pp. 36, May 2006.
[30] S. Gorsse and D.B. Miracle, "Mechanical properties of Ti-6Al-4V/TiB
composites with randomly oriented and aligned TiB reinforcements," Acta Materialia, vol. 51, no. 9, pp. 2427-2442, 23 May 2003.
[31] W. Lu, D. Zhang, X. Zhang, R. Wu, T. Sakata and H. Mori, "Microstructural
characterization of TiB in in-situ synthesized titanium matrix composites prepared by common casting technique," Journal of Alloys and Compounds, vol. 327 , no. 1, pp. 240-247, 30 August 2001.
[32] Y. Horiuchi, T. Inamura, H. Hosoda, K. Wakashima, H.Y. Kim and S. Miyazaki,
"Effect of boron addition on transformation behavior and tensile properties of Ti-Nb-Al alloy," Materials Science and Engineering A, vol. 498-440, pp. 830-834, 2006.
[33] S. Tamirisakandala, R.B. Bhat, D.B. Miracle, S. Boddapati, R. Bordia R. Vanover
and V.K. Vasudevan , "Effect of boron on the beta transus of Ti–6Al–4V alloy," Scripta Materialia , vol. 53, no. 2, pp. 217-222, July 2005.
[34] B. Cherukuri, R. Srinivasan and S. Tamarisakandala. Microstructural stability and
heat treatment of boron-modified beta titanium alloys. In Proceedings of International Symposium for Research Scholars, p. 51, Chennai, India, 2006
66
[35] J. Zhu, A. Kamiya , T. Yamada, W. Shi and K. Naganuma, "Influence of
boron addition on microstructure and mechanical properties of dental cast titanium alloys," Materials Science and Engineering A, vol. 339, no. 1, pp. 53-62, 2 January 2003.
[36] G.Y. Lai, High Temperature Corrosion of Engineering Alloys, Materials Park,
OH: ASM International, 1990. [37] D. Talbot and J. Talbot, Corrosion Science and Technology, Boca Raton: CRC
Press, 1998. [38] C. Xu and W. Gao, "Pilling-Bedworth ratio for oxidation of alloys,"
Materials Research Innovations, vol. 3, no. 4, pp. 231-235, March 2000.
[39] W. C. Carter, “Ellingham Diagrams,” [Online Document], 2002 Mar 12,
Available at: http://pruffle.mit.edu/3.00/Lecture_33_web/node4.html [40] M.R. Shanabarger, "Comparative study of the initial oxidation behavior of a
series of titanium–aluminum alloys," Applied Surface Science , vol. 134, no. 1-4, pp. 179-186, September 1998.
[41] B.V. Cockeram, "Growth and oxidation resistance of boron-modified and
germanium doped silicide diffusion coatings formed by the halide-activated pack cementation method," Surface and Coatings Technology, vol. 76-77, pp. 20-27, 1995.
[42] J. Unnam, R.N. Shenoy and Clark, “Effect of Alloy Chemistry and Exposure
Conditions on the Oxidation of Titanium”, NASA Technical Memorandum, National Aeronautic and Space Administration, Langley Research Center, Hampton VA. August 1984
[43] B.O. Aronsson, A. Krozer, J. Lausmaa, and B. Kasemo, "Commercially Pure
Titanium and Ti6Al4V: XPS Comparison Between Different Commercial Ti Dental Implants and Foils Prepared by Various Oxidation Procedures," Surface Science Spectra, vol. 4, no. 1, pp. 42-89, January 1996.
[44] H. Garbacz and M. Lewandowska, "Microstructural changes during oxidation of
titanium alloys," Materials Chemistry and Physics , vol. 81, pp. 542–547, 2003. [45] H. Guleryuz and H. Çimenoğlu, "Surface modification of a Ti–6Al–4V alloy by
thermal oxidation," Surface and Coatings Technology , vol. 192, no. 2-3, pp. 164-170, 21 March 2005.
[46] F.M. Güçlü; H. Çimenoğlu, E.S. Kayal, "The recrystallization and thermal
oxidation behavior of CP-titanium," Materials Science and Engineering C, vol. 26, no. 8, pp. 1367-1372, August 2006.
67
[47] I. Gurrappa, "An oxidation model for predicting the life of titanium alloy
components in gas turbine engines," Journal of Alloys and Compounds, vol. 389, no. 1-2, pp. 190-197, March 2005.
[48] H.L. Du, P.K. Datta, D.B. Lewis and J.S. Burnell-Gray, "Air Oxidation Behaviour
of Ti-6AL-4V Alloy between 650 and 850˚C," Corrosion Science, vol. 36, no. 4, pp. 631-642, 1994.
[49] M.N. Mungole, N. Singh and G.N. Mathur, "Oxidation behaviour of Ti6Al4V
titanium alloy in oxygen," Materials Science and Technology, vol. 18, no. 1, pp. 111-114, January 2002.
[50] S. Frangini, A. Mignone and F. de Riccardis, "Various aspects of the air oxidation
behaviour of a Ti6Al4V alloy at temperatures in the range 600–700 °C," Journal of Materials Science, vol. 29, no. 3, pp. 714-720, January 1994.
[51] E. Zhang, G. Zeng and S. Zeng , "Effect of in situ TiB short fibre on oxidation
behavior of Ti–6Al–1.2B alloy," Scripta Materialia, vol. 46, no. 11, pp. 811-816, 7 June 2002.
[52] Y. Qin, D. Zhang, W. Lu, J. Qin, and D. Xu, "Oxidation behavior of in situ
synthesized TiB/Ti-Al composite," Journal of Materials Science, vol. 40, no. 24, pp. 6553-6558, December 2005.
[53] D. B. Lee, M. H. Kim, C. W. Yang, S. H. Lee, M. H. Yang and Y. J. Kim, "The
Oxidation of TiB2 Particle-Reinforced TiAl Intermetallic Composites," Oxidation of Metals, vol. 56, no. 3-4, pp. 215-229, October 2001.
[54] Scandium version 5.0 Copyright Olympus Soft Imaging Systems GMBH
http://www.soft-imaging.net [55] Automation Creations Inc., MatWeb Material Property Data, [Online Database],
Copyright: 1996-2008, Available at: http://www.matweb.com. [56] B.G. Velasco and Pranesh Aswath, "Microstructural Stability, Microhardness and
Oxidation Behavior of In-Situ Reinforced Ti-8.5Al-1B-1Si (wt.%)," Journal of Materials Science, vol. 33, pp. 2203-2214, 1998.
APPENDIX A: Surface Area Measurements
Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 550 2 4.66 4.62 3.14 2.96 9.95 181.3386 b 550 2 4.68 4.68 2.62 2.485 10.03 168.9754 c 550 2 4.67 4.66 3.91 3.87 10.02 207.7361
2a 550 10 4.68 4.67 3.29 2.96 10.01 185.3764 b 550 10 4.64 4.65 3.33 3.36 10 190.8752 c 550 10 4.66 4.65 2.71 2.63 10.02 171.6511
3a 550 20 4.65 4.63 3.14 3.12 10 184.4466 b 550 20 4.68 4.65 4.67 4.66 10.02 230.4978 c 550 20 4.66 4.64 4.48 4.65 10 226.7528
4a 550 50 4.66 4.64 3.22 3.27 10.03 188.5517 b 550 50 4.65 4.65 4.18 4.18 10.04 216.1804 c 550 50 4.67 4.65 2.99 3.15 10.01 183.3654
5a 550 100 4.68 4.65 3.35 3.45 10 193.0205 b 550 100 4.73 4.75 3.75 3.72 10.02 205.2465 c 550 100 4.64 4.67 3.73 3.52 10.04 200.008
6a 550 200 4.62 4.65 3.85 3.61 10.01 202.0408 b 550 200 4.69 4.66 3.5 3.48 10.04 196.585
5006
c 550 200 4.67 4.66 4.64 4.65 10.03 230.0964
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 550 2 4.61 4.61 3.61 3.3 10.06 194.1229 b 550 2 4.61 4.6 3.55 3.55 10.01 195.9586 c 550 2 4.62 4.63 4.34 4.4 10.03 220.8625
2a 550 10 4.57 4.5 3.685 4.04 10.04 203.6423 b 550 10 4.59 5.57 3.3 3.28 10.03 201.3188 c 550 10 4.62 4.62 3.19 3.55 10.01 191.0986
3A 550 20 4.64 4.61 3.2 3.03 9.97 183.1519 b 550 20 4.67 4.66 3.88 3.59 10 202.849 c 550 20 4.61 4.61 3.78 3.87 10.04 204.6413
4a 550 50 4.63 4.58 3.13 3.14 10.04 184.2923 b 550 50 4.62 4.61 3.01 2.09 10 166.8411 c 550 50 4.62 4.62 3.48 3.58 10.02 195.9432
5a 550 100 4.21 4.4 4.64 4.64 10.05 219.7449 b 550 100 4.61 4.46 3.36 3.4 9.98 188.637 c 550 100 4.65 4.65 3.19 3.29 10.02 188.2476
6a 550 200 4.66 4.65 3.94 3.94 10.01 208.7533 b 550 200 4.65 4.64 3.53 3.5 10.01 196.0177
ISM
c 550 200 4.5 4.43 4.47 4.28 10.01 216.0522
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 550 2 4.65 4.61 3.18 3.18 10.03 186.1154 b 550 2 4.62 4.62 3.81 3.76 10 203.0734 c 550 2 4.63 4.62 3.53 3.72 10 198.5303
5011
2a 550 10 4.62 4.62 3.8 3.72 9.92 201.0016
A-1
b 550 10 4.64 4.62 3.34 3.36 9.99 190.4612 c 550 10 4.63 4.63 3.07 4 10.04 196.6873
3a 550 20 4.63 4.64 4.15 4.14 10.03 214.5509 b 550 20 4.64 4.63 2.62 2.81 9.99 172.0201 c 550 20 4.61 4.6 3.57 3.33 10.01 193.0368
4a 550 50 3.61 4.6 3.23 3.12 10.04 172.1947 b 550 50 4.61 4.59 2.69 2.8 9.99 172.006 c 550 50 4.61 4.62 3.6 3.55 9.99 196.6332
5a 550 100 4.65 4.66 3.55 3.53 9.98 196.5295 b 550 100 4.62 4.62 3.2 3.23 10.03 186.8767 c 550 100 4.59 4.6 2.9 2.96 9.98 177.126
6a 550 200 4.63 4.6 3.34 3.38 10.05 191.3097 b 550 200 4.62 4.61 3.75 3.44 10.03 197.876 c 550 200 4.62 4.64 2.19 2.19 10.02 156.9522
Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 550 2 6.33 6.31 4.47 4.45 10.03 272.6214 2a 550 10 6.34 6.34 3.7 3.7 9.98 247.3144 3a 550 20 6.35 6.36 3.49 3.5 10.03 242.0125 4a 550 50 6.35 6.38 4.55 4.57 10.06 277.8601 5a 550 100 6.36 6.33 3.5 3.5 10.02 241.7088 6a 550 200 6.35 6.35 3.99 4.01 10.02 258.214
Ti-64
Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 600 2 4.73 4.73 3.67 3.64 10.02 202.6117 b 600 2 4.24 4.24 4.58 4.63 10 215.9504 c 600 2 4.68 4.68 3.18 3.23 10.01 187.8565
2a 600 10 4.69 4.69 4.1 4.13 10 214.6987 b 600 10 4.68 4.68 3.57 3.46 10.04 197.456 c 600 10 4.29 4.27 4.51 4.58 10.06 216.4635
3a 600 20 4.7 4.7 2.9 3.05 10.02 181.772 b 600 20 4.66 4.65 3.33 3.125 10.03 188.172 c 600 20 4.64 4.64 3.88 3.88 10 206.4064
4a 600 50 4.66 4.66 3.28 3.65 10 194.7938 b 600 50 4.68 4.68 3.65 3.91 10 204.5808 c 600 50 4.65 4.65 3.4 3.4 10 192.62
5a 600 100 4.65 4.65 4.25 4.2 10 216.7925 b 600 100 4.66 4.66 3.35 3.4 10.06 193.1192 c 600 100 4.63 4.57 4.04 3.94 10 208.511
6a 600 200 4.2 3.8 4.36 4.45 10.05 204.1625 b 600 200 4.68 4.68 2.95 3.05 10.05 182.448
5006
c 600 200 4.65 4.65 3.75 3.76 10.03 203.5258
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 600 2 4.59 4.59 3.37 3.4 10.04 191.2123 b 600 2 4.59 4.6 3.43 3.31 10.02 190.5883 c 600 2 4.61 4.58 3.27 3.28 10.01 187.6545
2a 600 10 4.61 4.61 3.51 3.52 10.01 195.0708
ISM
b 600 10 4.6 4.6 3.73 3.65 10.02 200.0796
A-2
c 600 10 4.64 4.63 3.65 3.58 10.01 198.6764 3a 600 20 4.65 4.65 3.53 3.83 10 200.824 b 600 20 4.62 4.64 3.56 3.56 10.01 196.9294 c 600 20 4.6 4.6 3.33 3.33 10.02 189.5532
4a 600 50 4.59 4.59 3.29 3.28 10.04 188.2863 b 600 50 4.6 4.6 3.26 3.26 10.03 187.6636 c 600 50 4.61 4.59 3.65 3.36 10 194.3489
5a 600 100 4.6 4.6 3.17 3.05 10.03 183.2746 b 600 100 4.66 4.63 3.16 3.21 9.99 186.0313 c 600 100 4.61 4.6 3.77 3.81 10.03 203.3094
6a 600 200 4.6 4.6 3.5 3.56 10.05 195.889 b 600 200 4.6 4.61 3.45 3.42 10.01 192.597 c 600 200 4.62 4.62 3.03 2.99 10 180.4124
Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 600 2 4.64 4.61 2.96 2.96 10.02 179.3834 b 600 2 4.62 4.62 2.85 2.75 10.02 174.5688 c 600 2 4.6 4.61 3.4 3.42 10.02 192.0268
2a 600 10 4.6 4.61 3.3 3.21 10.02 187.4925 b 600 10 4.62 4.62 3.16 3.16 10.01 184.954 c 600 10 4.63 4.63 3.37 3.23 10.01 189.3166
3a 600 20 4.6 4.6 3.43 3.4 10.02 192.0386 b 600 20 4.61 4.61 3.68 3.76 10.05 201.7314 c 600 20 4.62 4.62 3.5 3.57 10.01 195.9265
4a 600 50 4.62 4.62 2.54 2.56 10 166.962 b 600 50 4.62 4.63 3.06 3.09 9.99 182.2899 c 600 50 4.6 4.6 3.77 3.65 10.03 200.8306
5a 600 100 4.65 4.66 3.62 3.62 10.01 199.3677 b 600 100 4.64 4.64 3.55 3.32 10.04 194.0228 c 600 100 4.66 4.66 3.29 3.28 10.01 189.6751
6a 600 200 4.66 4.63 3.12 3.11 10.03 184.6041 b 600 200 4.63 4.62 4.17 4.07 10 213.0105 c 600 200 4.63 4.64 3.51 3.53 10.01 195.8936
5011
A-3
Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 600 2 6.37 6.39 3.83 3.87 10 253.7264 2a 600 10 6.35 6.37 4.37 4.41 10.07 272.3462 3a 600 20 6.35 6.38 3.91 3.86 10.02 254.8653 4a 600 50 6.35 6.36 3.4 3.35 9.58 229.3228 5a 600 100 6.36 6.4 4.65 4.66 10.05 281.2015
Ti-64
6a 600 200 6.35 6.34 4.57 4.59 10.07 278.1496
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 700 2 4.78 4.57 4.65 4.63 10.03 230.245 b 700 2 4.63 4.6 3.56 3.66 10 197.8188 c 700 2 4.69 4.67 3.15 3.21 10 186.9642
2a 700 10 4.36 4.33 4.67 4.68 9.97 220.4844 b 700 10 4.65 4.64 4 3.82 9.98 207.0826 c 700 10 4.62 4.6 3.55 3.55 10 195.931
3a 700 20 4.65 4.64 3.76 3.76 10.01 203.1985 b 700 20 4.42 4.27 4.6 4.61 10 219.0167 c 700 20 4.63 4.63 3.47 3.48 10.02 194.6027
4a 700 50 4.68 4.66 3.67 3.66 9.98 200.5978 b 700 50 4.61 4.61 3.77 3.9 9.98 203.9209 c 700 50 4.7 4.7 4.66 4.64 10 230.71
5a 700 100 4.61 4.61 3.39 3.57 10.03 194.371 b 700 100 4.69 4.68 3.44 3.65 10.02 198.1448 c 700 100 4.66 4.64 3.67 3.6 10.01 199.6719
6a 700 200 4.65 4.65 4.12 3.6 10.01 206.2682 b 700 200 4.63 4.61 3.75 3.72 10.02 201.9459
5006
c 700 200 4.61 4.62 3.68 3.76 9.96 200.3692
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 700 2 4.61 4.61 3.28 3.35 9.99 188.9058 b 700 2 4.59 4.58 3.16 3.34 9.99 186.3449 c 700 2 4.6 4.58 2.85 2.93 10.01 176.279
2a 700 10 4.55 4.56 2.72 2.84 10.02 172.3198 2b 700 10 4.61 4.61 3.26 3.2 10 186.5806 c 700 10 4.64 4.62 3.3 3.29 10 189.0118
3A 700 20 4.6 4.61 3.78 3.43 9.97 196.9077 b 700 20 4.61 4.62 2.73 2.535 10 169.247 c 700 20 4.61 4.62 4.31 4.17 10 216.2345
4a 700 50 4.55 4.54 3.73 3.6 9.95 196.6945 b 700 50 4.61 4.62 3.39 3.34 9.99 190.4991 c 700 50 4.63 4.6 3.27 3.39 10.03 190.1108
5a 700 100 4.6 4.64 3.93 3.9 10.03 207.3861 b 700 100 4.64 4.62 3.12 3.15 10.04 184.951 c 700 100 4.6 4.6 3.13 3.1 10.02 183.2666
6a 700 200 4.62 4.58 3.01 3.32 9.97 183.9459 b 700 200 4.62 4.59 3.37 3.38 10 190.6836 c 700 200 4.57 4.57 3.57 3.67 10 196.8868
ISM
A-4
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1a 700 2 4.62 4.63 3.84 3.77 10 203.7959 b 700 2 4.6 4.58 3.41 3.32 10.01 190.1507 c 700 2 4.61 4.62 3.4 3.27 10 189.7814
2a 700 10 4.62 4.62 3.45 3.45 10.01 193.4394 b 700 10 4.6 4.6 3.53 3.29 9.99 191.4118 c 700 10 4.62 4.63 3.61 3.43 9.99 195.2962
3a 700 20 4.58 4.59 3.6 3.85 10.02 200.6919 b 700 20 4.6 4.58 3.22 3.12 10 184.3016 c 700 20 4.61 4.61 3.66 3.47 9.98 196.0423
4a 700 50 4.61 4.61 3.26 3.27 9.99 187.4458 b 700 50 4.6 3.58 3.48 3.44 10 179.3232 c 700 50 4.63 4.52 3.55 3.55 10.02 195.3075
5a 700 100 4.61 4.61 2.78 2.66 10 171.6784 b 700 100 4.63 4.65 3.26 3.35 10 189.5713 c 700 100 4.6 4.61 3.5 3.3 9.99 191.2529
6a 700 200 4.62 4.61 3.38 3.38 10 191.0974 b 700 200 4.6 4.59 3.49 3.52 10.02 194.5348
5011
c 700 200 4.62 4.6 3.5 3.5 10.05 195.281 Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1a 700 2 6.33 6.33 2.65 2.65 9.73 208.2998 2a 700 10 6.31 6.32 2.57 2.58 9.71 205.1661 3a 700 20 6.35 6.36 3.06 3.03 9.92 225.1978 4a 700 50 6.35 6.35 2.96 2.98 9.93 222.8142 5a 700 100 6.37 6.36 3.32 3.3 9.93 234.2819
Ti-64
6a 700 200 6.39 6.36 2.95 2.99 9.9 222.8979 Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1 850 2 3.72 3.37 2.5 2.47 9.43 131.3497 2 850 10 4.45 4.67 2.39 2.46 9.46 154.2799 3 850 20 4.59 4.61 2.42 2.39 9.32 152.6989 4 850 50 4.21 3.78 2.47 2.48 9.37 141.0209 5 850 100 4.32 4.44 2.4 2.43 9.4 148.9032
5006
6 850 200 4.43 3.89 2.3 2.25 9.32 138.8899 Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1 850 2 4.23 4.44 2.26 2.37 9.33 144.1716 2 850 10 3.52 3.72 2.22 2.2 9.31 124.553 3 850 20 4.2 4.2 2.48 2.5 9.48 147.7584 4 850 50 4.21 4.47 2.3 2.17 9.41 143.1244 5 850 100 4.58 3.94 2.24 2.26 9.47 142.463
ISM
6 850 200 3.44 3.74 2.3 2.31 9.39 127.2595 Sample
ID TºC time (h)
t1 (mm)
t2 (mm)
w1 (mm) w2 (mm) L (mm) Surface Area
(mm2) 1 850 2 4.35 4.57 2.24 2.28 9.4 146.4996 2 850 10 4.35 4.57 2.24 2.28 9.4 146.4996 3 850 20 4.27 4.35 2.44 2.42 9.45 148.3318 4 850 50 4.24 4.17 2.31 2.38 9.37 142.466 5 850 100 4.71 4.24 2.28 2.27 9.46 148.0736
5011
6 850 200 4.24 4.19 2.36 2.36 9.37 143.1103
A-5
Sample ID TºC time
(h) t1
(mm) t2
(mm) w1
(mm) w2 (mm) L (mm) Surface Area (mm2)
1 850 2 6.32 6.32 3.88 3.78 9.72 245.7272 2 850 10 6.35 6.35 3.77 3.75 9.83 246.5146 3 850 20 6.34 6.31 3.89 3.8 9.88 249.5998 4 850 50 6.3 6.3 0.385 3.87 9.76 191.3113 5 850 100 6.32 6.32 3.89 3.92 9.72 248.1332
Ti-64
6 850 200 6.33 6.33 3.7 3.7 9.85 244.433
A-6
APPENDIX B: Weight Measurements and Calculated Data
Sample
ID TºC time (h) Pre-ox
wt. (g) Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2 (g/mm2)2
0 0 5006-1 850 2 0.3685 0.3747 0.0062 131.3497 4.72E-05 2.23E-09 5006-3 850 10 0.4723 0.4894 0.0171 154.2799 0.000111 1.23E-08 5006-4 850 20 0.4555 0.4855 0.03 152.6989 0.000196 3.86E-08 5006-5 850 50 0.4123 0.4931 0.0808 141.0209 0.000573 3.28E-07 5006-6 850 100 0.441 0.5342 0.0932 148.9032 0.000626 3.92E-07
5006
5006-7 850 200 0.391 0.5458 0.1548 138.8899 0.001115 1.24E-06
Sample ID TºC time (h) Pre-ox
wt. (g) Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2 (g/mm2)2
0 0 ism1 850 2 0.4107 0.4177 0.007 144.1716 4.86E-05 2.36E-09 ism8 850 10 0.3377 0.3603 0.0226 124.553 0.000181 3.29E-08 ism2 850 20 0.4379 0.4688 0.0309 148.026 0.000209 4.36E-08 ism3 850 50 0.4132 0.4639 0.0507 140.3014 0.000361 1.31E-07 ism4 850 100 0.4 0.4786 0.0786 142.463 0.000552 3.04E-07
ism
ism9 850 200 0.3527 0.4747 0.122 127.2595 0.000959 9.19E-07
Sample ID TºC time (h) Pre-ox
wt. (g) Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2 (g/mm2)2
0 0
5011-1 850 2 0.4214 0.4298 0.0084 146.4996 5.73E-05 3.29E-09
5011-1 850 10 0.4214 0.4639 0.0425 146.4996 0.00029 8.42E-08
5011-5 850 20 0.413 0.4666 0.0536 148.3318 0.000361 1.31E-07 5011-2 850 50 0.4073 0.4918 0.0845 142.466 0.000593 3.52E-07 5011-3 850 100 0.4185 0.5204 0.1019 148.0736 0.000688 4.74E-07
5011
5011-4 850 200 0.4145 0.5405 0.126 143.1103 0.00088 7.75E-07
Sample ID TºC time (h) Pre-ox
wt. (g) Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2 (g/mm2)2
0 0 1 850 2 1.0696 1.0775 0.0079 245.7272 3.21E-05 1.03E-09 2 850 10 1.04 1.0677 0.0277 246.5146 0.000112 1.26E-08 3 850 20 1.0755 1.1154 0.0399 249.5998 0.00016 2.56E-08 4 850 50 1.0914 1.1703 0.0789 191.3113 0.000412 1.7E-07 5 850 100 1.0984 1.2138 0.1154 248.1332 0.000465 2.16E-07
Ti-64
6 850 200 1.0562 1.2102 0.154 244.433 0.00063 3.97E-07
Sample ID TºC time (h) Pre-ox
wt. (g) Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2 (g/mm2)2
1A 700 2 0.9835 0.9842 0.0007 230.25 3.04E-06 9.24E-12 b 700 2 0.7579 0.7585 0.0006 197.82 3.03E-06 9.20E-12 c 700 2 0.6851 0.6857 0.0006 186.96 3.21E-06 1.03E-11
2a 700 10 0.9233 0.9256 0.0023 220.48 1.04E-05 1.09E-10 b 700 10 0.8229 0.8249 0.002 207.08 9.66E-06 9.33E-11
5006
c 700 10 0.749 0.7514 0.0024 195.93 1.22E-05 1.50E-10
B-1
3a 700 20 0.7986 0.8019 0.0033 203.20 1.62E-05 2.64E-10 b 700 20 0.9065 0.9097 0.0032 219.02 1.46E-05 2.13E-10 c 700 20 0.7389 0.7405 0.0016 194.60 8.22E-06 6.76E-11
4a 700 50 0.7763 0.7821 0.0058 200.60 2.89E-05 8.36E-10 b 700 50 0.8081 0.8163 0.0082 203.92 4.02E-05 1.62E-09 c 700 50 0.9972 1.0078 0.0106 230.71 4.59E-05 2.11E-09
5a 700 100 0.7323 0.742 0.0097 194.37 4.99E-05 2.49E-09 b 700 100 0.7571 0.7648 0.0077 198.14 3.89E-05 1.51E-09 c 700 100 0.7711 0.7794 0.0083 199.67 4.16E-05 1.73E-09
6a 700 200 0.8053 0.8135 0.0082 206.27 3.98E-05 1.58E-09 b 700 200 0.7933 0.8034 0.0101 201.95 5.00E-05 2.50E-09 c 700 200 0.7798 0.7948 0.015 200.37 7.49E-05 5.60E-09
Sample
ID TºC time (h)
Pre-ox wt. (g)
Post-ox wt. (g) ∆W (g) A (mm2) g/mm2
(∆W/A)2
(g/mm2)2
1a 700 2 0.6965 0.6971 0.0006 188.91 3.18E-06 1.01E-11 b 700 2 0.6794 0.6802 0.0008 186.34 4.29E-06 1.84E-11 c 700 2 0.6136 0.6142 0.0006 176.28 3.40E-06 1.16E-11
2a 700 10 0.5787 0.5805 0.0018 172.32 1.04E-05 1.09E-10 b 700 10 0.6767 0.6793 0.0026 186.58 1.39E-05 1.94E-10 c 700 10 0.6999 0.7019 0.002 189.01 1.06E-05 1.12E-10
3a 700 20 0.7536 0.7564 0.0028 196.91 1.42E-05 2.02E-10 b 700 20 0.558 0.5601 0.0021 169.25 1.24E-05 1.54E-10 c 700 20 0.8911 0.8933 0.0022 216.23 1.02E-05 1.04E-10
4a 700 50 0.761 0.7661 0.0051 196.69 2.59E-05 6.72E-10 b 700 50 0.7156 0.7211 0.0055 190.50 2.89E-05 8.34E-10 c 700 50 0.701 0.7078 0.0068 190.11 3.58E-05 1.28E-09
5a 700 100 0.8205 0.8295 0.009 207.39 4.34E-05 1.88E-09 b 700 100 0.6577 0.6649 0.0072 184.95 3.89E-05 1.52E-09 c 700 100 0.6562 0.6647 0.0085 183.27 4.64E-05 2.15E-09
6a 700 200 0.6643 0.6704 0.0061 183.95 3.32E-05 1.10E-09 b 700 200 0.7079 0.7168 0.0089 190.68 4.67E-05 2.18E-09
ISM
c 700 200 0.755 0.7662 0.0112 196.89 5.69E-05 3.24E-09
B-2
APPENDIX C: Graphical data
Comparative gain W/A vs time T=550°C
0.00E+00
5.00E-07
1.00E-06
1.50E-06
2.00E-06
2.50E-06
3.00E-06
3.50E-06
4.00E-06
4.50E-06
0 20 40 60 80 100 120 140 160 180 200
time (h)
W/A
(g/m
m2 )
5006ISM5011Ti-64
Figure 1: Normalized weight gain over time at 550ºC
Comparative W/A vs time T=600°C
0
0.000002
0.000004
0.000006
0.000008
0.00001
0.000012
0 50 100 150 200 250
time (h)
W/A
(g/m
m2 )
5006ISM5011Ti-64
Figure 2: Normalized weight gain over time at 600ºC
C-1
Comparison W/A vs time T=700°C
0
0.00001
0.00002
0.00003
0.00004
0.00005
0.00006
0.00007
0.00008
0 20 40 60 80 100 120 140 160 180 200
time (h)
W/A
(g/m
m2 )
5006ISM5011Ti-64
Figure 3: Normalized weight gain over time at 700ºC
Comparison W/A vs time T=850°C
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 50 100 150 200 250
time (h)
W/A
(g/m
m2 )
5006ISM5011Ti-64
Figure 4: Normalized weight gain over time at 850ºC
C-2
Kp T=550°C
y = 7.08E-14x (Ti-64)
y = 3.35E-14x (5011)
y = 3.17E-14x (5006)
y = 4.20E-14x (ISM)
0
2E-12
4E-12
6E-12
8E-12
1E-11
1.2E-11
1.4E-11
1.6E-11
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
5006ISM5011Ti-64
Figure 5: Normalized weight gain2 as a function of time at 550ºC
Kp T=600°C
y = 4.66E-13x (Ti-64)
y = 1.44E-13x (ISM)
y = 2.23E-13x (5011)
y = 1.90E-13x (5006)
0
1E-11
2E-11
3E-11
4E-11
5E-11
6E-11
7E-11
8E-11
9E-11
1E-10
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
5006ISM5011Ti-64
Figure 6: Normalized weight gain2 as a function of time at 600ºC
C-3
Kp T= 700°C
y = 2.04E-11x (Ti-64)
y = 1.80E-11x (5006)
y = 1.44E-11x (5011)
y = 1.27E-11x (ISM)
0.0E+00
1.0E-09
2.0E-09
3.0E-09
4.0E-09
5.0E-09
6.0E-09
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
5006ISM5011Ti-64
Figure 7: Normalized weight gain2 as a function of time at 700ºC
Kp T=850°C
y = 4.22E-09x (5011)
y = 4.19E-09x (ISM)
y = 5.75E-09x (5006)
y = 2.08E-09x (Ti-64)
0
0.0000002
0.0000004
0.0000006
0.0000008
0.000001
0.0000012
0.0000014
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
5006ISM5011Ti-64
Figure 8: Normalized weight gain2 as a function of time at 850ºC
C-4
Kp (5006)
y = 3.17E-14x (550°C)
y = 1.90E-13x (600°C)
0
1E-12
2E-12
3E-12
4E-12
5E-12
6E-12
7E-12
8E-12
9E-12
1E-11
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
0
1E-11
2E-11
3E-11
4E-11
5E-11
6E-11
550°C600°C
Figure 9: 5006: Kp at 550ºC and 600ºC
Kp (5006)
y = 5.75E-09x (850°C)
y = 1.80E-11x (700°C)
0.0E+00
2.0E-07
4.0E-07
6.0E-07
8.0E-07
1.0E-06
1.2E-06
1.4E-06
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
0.0E+00
1.0E-09
2.0E-09
3.0E-09
4.0E-09
5.0E-09
6.0E-09
850°C700°C
Figure 10: 5006: Kp at 700ºC and 850ºC
C-5
Kp (ISM)
y = 1.440E-13x (600°C)
y = 4.202E-14x (550°C)
0
5E-12
1E-11
1.5E-11
2E-11
2.5E-11
3E-11
3.5E-11
4E-11
4.5E-11
5E-11
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
550°C600°C
Figure 11: ISM: Kp at 550ºC and 600ºC
Kp (ISM)
y = 1.27E-11x (700°C)
y = 4.19E-09x
0.0E+00
5.0E-10
1.0E-09
1.5E-09
2.0E-09
2.5E-09
3.0E-09
3.5E-09
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
0.E+00
1.E-07
2.E-07
3.E-07
4.E-07
5.E-07
6.E-07
7.E-07
8.E-07
9.E-07
1.E-06
700°C850°C
Figure 12: ISM: Kp at 700ºC and 850ºC
C-6
Kp (5011)
y = 2.23E-13x (550°C)
y = 3.35E-14x (600°C)
0
2E-12
4E-12
6E-12
8E-12
1E-11
1.2E-11
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
0
1E-11
2E-11
3E-11
4E-11
5E-11
6E-11
7E-11
8E-11
550°C600°C
Figure 13: 5011: Kp at 550ºC and 600ºC
Kp (5011)
y = 1.44E-11x (700°C)y = 4.22E-09x (850°C)
0.0E+00
5.0E-10
1.0E-09
1.5E-09
2.0E-09
2.5E-09
3.0E-09
3.5E-09
0 50 100 150 200 250
time (h)
(W/A
)2 (g
/mm
2 )2
0.0E+00
1.0E-07
2.0E-07
3.0E-07
4.0E-07
5.0E-07
6.0E-07
7.0E-07
8.0E-07
9.0E-07
700°C850°C
Figure 14: 5011: Kp at 700ºC and 850ºC
C-7
Kp Ti-64
y = 4.656E-13x
y = 7.085E-14x
0
1E-11
2E-11
3E-11
4E-11
5E-11
6E-11
7E-11
8E-11
9E-11
1E-10
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2 550°C
600°C
Figure 15: Ti-64: Kp at 550ºC and 600ºC
Kp Ti-64
y = 2E-11x + 2E-10 y = 2E-09x + 7E-09
0.0E+00
5.0E-10
1.0E-09
1.5E-09
2.0E-09
2.5E-09
3.0E-09
3.5E-09
4.0E-09
4.5E-09
0 50 100 150 200 250
time (h)
(W/A
)2 (g/m
m2 )2
0.E+00
5.E-08
1.E-07
2.E-07
2.E-07
3.E-07
3.E-07
4.E-07
4.E-07
5.E-07
700°C850°C
Figure 16: Ti-64: Kp at 700ºC and 850ºC
C-8
Kp vs 1/T (5006)
y = -1.6443x + 6.2878
-14
-13
-12
-11
-10
-9
-88 8.5 9 9.5 10 10.5 11 11.5 12 12.5 13
104/T (K-1)
Kp
Figure 17: 5006: Kp vs 1/T
Kp vs 1/T (5011)
y = -1.5836x + 5.6037
-14
-13
-12
-11
-10
-9
-88 9 10 11 12 13 14
104/T (K-1)
Kp (
g2 /mm
4 s)
Figure 18: 5011: Kp vs 1/T
C-9
Kp vs 1/T (ISM)
y = -1.5791x + 5.5183
-14
-13
-12
-11
-10
-9
-88 8.5 9 9.5 10 10.5 11 11.5 12 12.5
104/T (K-1)
Kp (
g2 /mm
4 s)
Figure 19: ISM: Kp vs 1/T
Kp vs 1/T (Ti-64)
y = -1.3856x + 3.6085
-14
-13
-12
-11
-10
-9
-88 8.5 9 9.5 10 10.5 11 11.5 12 12.5
104/T (K-1)
Kp
(g2 /m
m4 s)
Figure 20: Ti-64: Kp vs 1/T
C-10
900µm
Figure 21: Composition of ISM at 550˚C
900µm
Figure 22: Composition of Ti-64 at 550˚C
C-11
900µm
Figure 23: Composition of 5006 at 600˚C
900µm
Figure 24: Composition of 5011 at 600˚C
C-12
900µm
Figure 25: Composition of ISM at 600˚C
900µm
Figure 26: Composition of Ti-64 at 600˚C
C-13
900µm
Figure 27: Composition of 5006 at 700˚
900µm
Figure 28: Composition of 5011 at 700˚
C-14
900µm
Figure 29: Composition of ISM at 700˚
900µm
Figure 30: Composition of Ti-64 at 700˚
C-15
Ti-64 550°C 10h 100g
y = 0.0x + 348.9
200
220
240
260
280
300
320
340
360
380
400
0 500 1000 1500 2000 2500 3000 3500
distance from edge (mm)
hard
ness
(kno
op)
Figure 31: Hardness Test Results
C-16
APPENDIX D: Micrographs
Figure 1: Oxide scale of 5006 at 550˚C
Figure 2: Oxide scale of 5011 at 550˚C
Figure 3: Oxide scale of ISM at 550˚C
Figure 4: Oxide scale of Ti-64 at 550˚C
Figure 5: Oxide scale of 5006 at 600˚C
Figure 6: Oxide scale of 5011 at 600˚C
D-1
Figure 7: Oxide scale of ISM at 600˚C
Figure 8: Oxide scale of Ti-64 at 600˚C
Figure 9: Oxide scale of 5006 at 700˚C
Figure 10: Oxide scale of 5011 at 700˚C
Figure 11: Oxide scale of ISM at 700˚C
Figure 12: Oxide scale of Ti-64 at 700˚C
D-2
Figure 13: Oxide scale of 5006 at 850˚C (optical mag. 10x)
Figure 14: Oxide scale of 5011 at 850˚C
(optical mag. 10x)
Figure 15: Oxide scale of ISM at 850˚C (optical mag. 10x)
Figure 16: Oxide scale of Ti-64 at 850˚C (optical mag. 10x)
D-3