Development of High-Strength ODS Steels for

Nuclear Energy Applications

D.T. Hoelzer, J. Bentley, M.K. Miller, M.K. Sokolov and T.S. ByunOak Ridge National Laboratory, USA

M. LiArgonne National Laboratory, USA

ODS 2010 Materials Workshop

Qualcomm Conference CenterJacobs Hall

University of California, San DiegoNov. 17-18, 2010

Nuclear power’s proven performance

Source: Energy Information AdministrationNuclear Regulatory Commission

92%

Increases in capacity factor at US nuclear plants in the last 15 years is equal to building 26 new 1,000-MW plants

Currently providing 16% of the world’s electricity

Dramatic reduction in stress corrosion cracking issues in steam generators, etc.

Average capacity factor for US nuclear power plants has exceeded 90% for the past decade

Irradiation Produces Defect Microstructures

Irradiation Temperature (T/TM)0.2 0.3 0.4 0.5 0.6 0.7

Dislocation loops Voids, precipitates, solute segregation

Grain boundary helium cavities

Radiation Damage Can Produce Large Changes in Structural Materials

• Radiation hardening and embrittlement (<0.4 TM, >0.1 dpa)

• Phase instabilities from radiation-induced precipitation(0.3-0.6 TM, >10 dpa)

• Irradiation creep(<0.45 TM, >10 dpa)

• Volumetric swelling from void formation(0.3-0.6 TM, >10 dpa)

• High temperature He embrittlement(>0.5 TM, >10 dpa)

0

200

400

600

800

1000

1200

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

USJF82Hss2

Eng

inee

ring

Stre

ss, M

Pa

Engineering Strain, mm/mm

Unirradiated YS

200°C/10 dpa250°C/3 dpa

300°C/8 dpa

400°C/10 dpa

400°C/34 dpa

500°C/8 dpa

500°C/34 dpa

600°C/8 dpa

Tirr

=Ttest

100 nm

50 nm50 nm

100 nm100 nmafter S.J. Zinkle, Phys. Plasmas 12 (2005) 058101

Gen IV Fission and Fusion Reactors Require the Development of Higher Temperature Materials with Adequate Radiation Resistance

Fission(Gen. I)

Fission(Gen. IV)

Fusion (Demo)

Structural alloy maximum temperature

<300˚C 500-1000˚C 550-1000˚C

Max dose for core internal structures

~1 dpa ~30-150 dpa ~200 dpa

Max transmutation helium concentration

~0.1 appm ~3-15 appm ~2000 appm(~10000 appm for

SiC)Coolants H2O He, H2O, Pb-Bi,

NaHe, Pb-Li, Li

Structural Materials Zircaloy, stainless steel

Ferritic steel, SS, superalloys, C-

composite

ODS &Ferritic/ martensitic steel, V

alloy, SiC composite

• High sink strength from internal interfaces - trap vacancies to enhance recombination with self-

interstitial atoms- trap He to suppress bubble formation on grain

boundaries

• Desirable combination of mechanical properties- elevated temperature strength and creep properties- low temperature fracture toughness

• Achievable with second phase dispersions- nano-size particles- high number density- thermally stable

Concept for Developing Radiation Tolerant Materials

Discovery of Oxygen-Enriched Nanoclusters (NC) in Ferritic Alloys Show Promise for Achieving this Goal

Radius: 2.0 ± 0.8 nmNumber density: 1.4 x 1024 m-3

Composition: 8.1 ± 5.2 %Y42.1 ± 5.6 %Ti44.4 ± 8.2 %O

Characteristics

• Background- Nanoclusters first observed in 12YWT (ORNL)

• Mechanically alloyed Fe - 12%Cr - 3%W - 0.4%Ti- 0.25%Y2O3 ferritic alloy

• Developed in Japan in late 1990’s by Kobe Steel (T. Okuda) and Nagoya University (K. Miyahara and I.S. Kim)

- Similar oxygen-enriched NC observed in MA957• Patented in 1978 by INCO

Mitigating Neutron Irradiation Damage

He

n

n

i

v

n NC

v

NFAB

L

FMS

Dislocation

GB

NFHe bubble

SIA recombines

trapped vacancy

NFHe bubble

SIA recombines

trapped vacancy

NC

Thanks to Bob Odette, UCSB

Nanoclusters have a high sink strength and trap (getter) both He (in fine bubbles) and vacancies (to enhance self-healing of damage by recombination with self-interstitial atoms, SIA)

Proof of Concept: Results of In-Situ Neutron + He Implantation of MA957 (JP26: 9 dpa at 500°C in HFIR)

specimen substrate

nth

nth

αα

59Ni

58Ni + nth → 59Ni + γ59Ni + nth → 56Fe + 4He (4.76 MeV)

“The transport and fate of helium in nanostructured ferritic alloys at fusion relevant He/dpa ratios and dpa rates”T. Yamamoto, G.R. Odette, P. Miao, D.T. Hoelzer, J. Bentley, N. Hashimoto, H. Tanigawa, R.J. KurtzJ Nucl Mater 367-370 (2007) 399-412

4-µm thick NiAl layer generates ~380 appm He in adjacent 6 µm of MA957

FIB Lift-Out Specimen

• Compared to Eurofer 97, the results indicate that NC play an important role in mitigating the accumulation of point defects as well as trapping He in ~1-2 nm bubbles

EFTEM Fe M Jump Ratio

Survival of NC and He Trapping in MA957-512 nm Under Focus

Eurofer 97

Development of Advanced ODS 14YWT Ferritic Alloy

• A major focus was to reduce the grain size– increase the strength and improve other properties, i.e. fracture– increase the sink strength for radiation tolerance

• The development of 14YWT began in 2001 since MA957 was discontinued by INCO and 12YWT was produced once

• Numerous small heats have been produced over 9 yearsHeat Extrusion Date Mass (g) Temp. (ºC) Dispersion History

OE14YWT-SM1 29-Apr-02 200 850 NC Bi-Modal GS: SANSOE14YWT-SM2 08-Feb-04 200 850 NC Poor milling conditionOE14YWT-SM3 17-Mar-04 200 850 NC SANS; Pb and super critical water corrosion testsOE14YWT-SM4 18-Mar-04 200 850 NC Tensile test; Steam corrosion testOE14YWT-SM5 17-Sep-04 200 850 NC Navy ATR irradiation experiementOE14YWT-SM6 28-Jan-05 800 850 NC Fracture toughness, tensile; HFIR Rabbit; Matrix IOE14YWT-SM7 03-Aug-05 1000 850 NC Navy ATR irradiation experiementOE14YWT-SM8 03-Aug-05 1000 850 NC Given to BES ProjectOE14YWT-SM9 03-Aug-05 200 850 NC

OE14YWT-SM10 27-Jun-07 1200 850 NC INERI testing; Matrix II

OE14YWT-CR1 18-Sep-01 200 1175 OxideOE14YWT-CR2 Fall, 2001 200 1175 Oxide Tensile testOE14YWT-CR3 29-Apr-02 200 850 NC Bi-Modal GS; SINQOE14YWT-CR4 17-Mar-04 200 850 NC Tensile test; SANSOE14YWT-CR5 17-Sep-04 200 850 NC Coarse particle study; Bi-Modal GS

10 h milling test

High temperature extrusion

Atomized Fe-alloy powder is ball milled with Y2O3 powderThe NFA 14YWT is produced by Mechanical Alloying

Y2O3

(Fe-14Cr-3W-0.4Ti)

Watson-Stillman Co. (Roselle, NJ) unit with total capacity of 1,250 tons (2.5 x 106 lbf)

CM08CM08 CM01CM01Zoz Attritor MillsZoz Attritor Mills

Ball milling is critical in mechanical alloying (MA)

As-Received Milled 5h Milled 80h

• Spherical particles undergo extensive plastic deformation during early stages of ball milling and rapidly develop into large mm size flakes

• Flakes become brittle due to work hardening and simply fracture into smaller particles during latter stages of ball milling

Up to 5 µm Y2O3 agglomerateson surface of Fe alloy sphere

(1) Must uniformly disperse Y2O3 in each Fe alloy particle(2) Must force Y2O3 into solution in the bcc Fe lattice

When ball milling does not distribute Y effectively

Poor milling conditionsY not dispersed uniformly

Optimized milling conditionsY dispersed uniformly

Uniform grain size (<500 nm)Homogeneous NC distribution

14YWT ball milled for 40 h and extruded at 850ºC

Bi-Modal grain sizeNC present only in Y-rich regions

Other factors to consider: Starting powder size and milling time

Y2O3 can be forced into solution with bcc Fe by milling

• Atom maps revealed no evidence of nanoclusters or remnants of the Y2O3particles in the 40 h ball milled powder of 14YWT

• Significantly higher O level in the bcc Fe matrix compared to equilibrium level

• The formation of NC occurs during annealing or consolidation of the ball milled powders at high temperatures

Local Electrode Atom Probe

Optimum ball milling produces nano-size grains and high concentration of nanoclusters in 14YWT

Grain size = 136 (+/- 14) nmGAR = ~ 1.2

BF TEM of SM10 Heat Local Electrode Atom Probe (LEAP)

• Very high interfacial surface area (NC and grain boundaries) may enhance trapping and recombination of point defects

O YTi CrFe

Nv = 1-7 x 1023 m-3 <r> = 1.8 +/- 0.4 nm

Grain boundaries

Nucleation of NC (and larger oxide particles) on grain boundaries stabilizes the nano-size grains

Local Electrode Atom Probe (LEAP)EFTEM Fe M Jump Ratio Image

Tempered Martensitic 9Cr Steel

0

200

400

600

800

1000

1200

1400

1600

1800

0 100 200 300 400 500 600 700 800 900

YS - 14YWTUTS - 14YWTYS - 12YWTUTS - 12YWTYS - MA957UTS - MA9570.2% PS - PM2000UTS - PM2000YS - 9Cr-2WVTaUTS - 9Cr-2WVTa

Stre

ss (M

Pa)

Temperature (ºC)

NFA (Contain NC)

* * **

Tested at ORNL ( 10-3s-1)Hamilton et al., PNNL-13168, 2000 ( 4.1 x 10-4s-1)

ε = ε = .

.

*

*

*

*

ODS

High temperature strength of 14YWT (SM4 heat) is similar to 12YWT and MA957

0100200300400500600700800900

10001100120013001400150016001700

0 10 20 30 40 50 60

22ºC364ºC500ºC600ºC650ºC700ºC800ºC

Stre

ss (M

Pa)

Crosshead Displacment (mm/mm)

High-strength properties of 14YWT heats is offset with low ductility properties

0200

400

600800

100012001400

1600

1800

20002200

2400

0 5 10 15 20 25

-196ºC-150ºC-100ºC25ºC300ºC500ºC600ºC700ºC

Stre

ss (M

Pa)

Strain (%)

14YWT-SM4 Heat 14YWT-SM6 Heat

• Low uniform elongation, but extensive plastic deformation occurs after plastic instability until failure

• Total elongation initially increases with higher temperatures, but then starts to decrease up to 800ºC

0

50

100

150

200

250

300

350

400

450

22000 24000 26000 28000 30000 32000 34000 36000

NFA 12YWTNFA MA9579Cr-WMoVNb SteelNFA 14YWT

LMP T(K)[25 + log10t(h)]

Stre

ss (M

Pa)

900°C, 1104h

800°C, 14235h

800°C, 817h

650°C, 13000h

600°C, 17000h

650°C, 1080h650°C16h

650°C>72h

825°C, 63480h(in test)

800°C, 38555h

800°C, 22464h(in test)

Creep performance of NFA 12YWT, MA957 and 14YWT

• Creep test on ruptured MA957 started in Oct., 2003 (INERI)- Extensometer creep strain prior to rupture was 0.361% (0.003 in. displacement)- The minimum creep rate was ~1.2 x 10-11 s-1 (dε/dt)

• Creep test on 14YWT-SM10 started in April, 2008

• Klueh et al, J. Nucl. Mat., (2005)• I-NERI FY01-04

-1.000

0.000

1.000

2.000

3.000

4.000

5.000

6.000

7.000

0 10000 20000 30000 40000 50000

TIME (HOURS)

CREE

P S

TRA

IN (%

)

Ruptured after ~38,555 h

NC are very stable at elevated temperatures

• EFTEM revealed no significant change in the size of the NC after 38,555 h (~4.4 years) at 800ºC and 100 MPa

• LEAP analysis has confirmed these results

After 38,555 h at 800ºCAs-received

Fe M jump ratioFe M jump ratio

3.6 nm3.6 nm

Fe M jump ratioFe M jump ratio

MA957 Creep Specimen

Creep tests initiated on 14YWT (SM10 heat) in 2009 to study creep deformation mechanisms

• Tests conducted using MTS with load control and extensometer

• Stress exponent = 24 for 14YWT at 800ºC

• Results are consistent with those for MA957 tested at 600 to 700ºC1

that showed a high stress exponent (n ~ 35) and activation energy (815 kJ/mol) for creep

• High stress exponents are consistent with threshold stress mechanism

10-11

10-10

10-9

10-8

10-7

10-6

10-5

10-4

10-3

0 100 200 300 400 500

600ºC - MA957650ºC - MA957700ºC - MA957800ºC - MA957875ºC - MA957900ºC - MA957925ºC - MA957

650ºC - 12YWT700ºC - 12YWT750ºC - 12YWT800ºC - 12YWT900ºC - 12YWT800ºC - 14YWT

Min

imum

Cre

ep R

ate

(s-1

)

Stress (MPa)

***

1 B. Wilshire and T.D. Lieu, Mat. Sci, Engin. A, 386, (2004), 81.

• Initial results suggest that the nano-size grain structure of 14YWT-SM10 does not significantly degrade the creep properties

Fracture toughness transition temperature of 14YWT

• FTTT in the L-T orientation is very low, but depends on the 14YWT heat► grain size difference is not a major factor

• FTTT for 14YWT heats were still significantly lower than that of 12YWT

• New results for SM10 showed T0 of -84ºC in L-T compared to 18ºC in T-L► anisotropy still had a significant effect on the FTTT

Two heats of 14YWT alloys

TEMPERATURE, oC

-200 -100 0 100 200 300 400 5001-T

FRA

CTU

RE

TOU

GH

NES

S, M

Pam

1/2

0

50

100

150

200

250

12YWTTo=102oC

K from JIc

14YWT-SM6To<-150oC

14YWT-SM10To=-84oC

SM6GS: 387 +/- 80 nm

SM10GS: 136 +/- 14 nm

L-T Orientation

Fracture toughness of 14YWT (SM6) after low dose and temperature HFIR neutron irradiation experiment

-250 -200 -150 -100 -50 0 50 100 150 200 250 30025

50

75

100

125

150

175

200Fr

actu

re T

ough

ness

[MP

a√m

Temperature (ºC)

KJc(1T) (unirr.)KJ1c (unirr.)KJc(1T) (Tirr. = 300ºC)KJ1c (Tirr. = 300ºC)

No shift in FTTT observed after irradiation at 300ºC to ~ 1.5 dpa

Model requires high concentration of vacancies that only ball milling can produce at low temperatures

-5

-4

-3

-2

-1

0

0 1 2 3 4 5 6Number of Ti atoms

Oi

Oi-vacancy pair

FeTiO

• Y+Ti (M):O ratio ≅ 1.2 (not consistent with known oxides)• Unusually strong vacancy-oxygen binding in bcc Fe• But, very high vacancy formation energy

Theoretical model indicates vacancies may play a vital role in the formation and stability of nanoclusters

TiO2 (-4.3 eV)

O in defect-free Fevacancies

Ti

YB

indi

ng e

nerg

y of

O (e

V/O

)Stable

Nanoclusters

Os-Os interactions

O-vacancy pair (Os)

Ti:O (~1:1) clusters

(-0.6 eV)

Fu and Krcmar; First Principles Theory (BES: C.T. Liu, P.I.)

Ti (and Y) strengthen O-V binding energy

Summary of R&D of advanced ODS ferritic alloys over the past 9 years

• Advanced ODS ferritic alloys strengthened by nanoclusters possess attractive high-temperature strength and creep properties

– But ductility and failure mechanisms suffer

• Nanoclusters show remarkable stability during long-term creep; low (neutron) and high (ion) dose irradiations; and short-term high-temperature exposures

– NC appear to trap He and possibly point defects, causing their recombination

– But information about their structure and chemistry is still lacking

• Development of 14YWT has approached “maturity”– Numerous small heats have been produced for a variety of studies

• But:– heat-to-heat variations still influence mechanical properties– Plus, anisotropy still affects the mechanical properties, even though

the GAR is reduced with nano-size grains

Research sponsorship

• Laboratory Directed Research and Development, ORNL

• Office of Nuclear Energy, Science and Technology (FY01-04 INERI; FY07-09 INERI; FY09 GNEP; and FY10+ FCRD)

• Office of Fusion Energy Science

• Research at the Oak Ridge National Laboratory SHaRE User Facility is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy

M.K. Miller, J. Bentley, C.L. Fu, M.A. Sokolov, D.A. McClintockOak Ridge National Laboratory, TN

S.A. MaloyLos Alamos National Laboratory

G.R. Odette, T. Yamamoto University of California, Santa Barbara, CA

M.J. Alinger, GE Global Research, NY

B. WirthUniversity of California, Berkeley, CA (Now at University of Tennessee, Knoxville

M. LiArgonne National Laboratory, IL

Contributors