Cchemistry materials science Deformation of Single Crystal...
Transcript of Cchemistry materials science Deformation of Single Crystal...
Deformation of Single Crystal Copper Under Extreme Conditions – Effect of Loading PathJ. M. McNaney1, B. Torralva1, E. M. Bringa1, M. S. Schneider2, J. S. Harper1, K. T. Lorenz1, B. A. Remington1, M. A. Meyers2
1Lawrence Livermore National Laboratory, 2University of California, San Diego
CMSchemistry &materialsscienceCMSchemistry &materialsscience
Performance Measure 2.1, 2.3, 2.6, 5.1, 5.2 and 5.4
CMS-05-66 This work was funded by LLNL Laboraory Directed Research and Development (LDRD)03-ERD-015
Material behavior at high pressure andstrain rate is not well understood and
is programmatically significant
A laser based platform is used toaccess shockless loading paths
Hydrodynamic simulations identifymaterial locations to be examined
*This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under Contract No. W-7405-Eng-48.
Knowledge of material behavior along loading paths off of theshock Hugoniot is needed to understand the physical processes
involved as well as constrain the form of the models used todescribe them.
Localized sample excision providesmaterial for comparing along
different loading paths
Under shock loading, the trans ition from dislocation cells tostacking faults occurs over the range 8-24 GPa. The incompletestacking fault formation at 20 GPa indicates it is very near thethreshold. Under shockless loading no transition is seen over thesame pressure range.
Transmission electron microscopy samples are removed from therecovered material using a focused ion beam method. This methodallows us to precisely match the location of the sample to thedesired loading conditions.
Samples deformed under shockless loading display a gradualtransition over the 8-42 GPa pressure range. This result, the first ofits kind, is generally supportive of the MD simulations.
Significant changes in deformationbetween loading paths is observed
Summary• This work represents the first study looking at high pressure, high strain rate deformation mechanisms occurring near the isentrope• Our observations confirm that significant differences in behavior exist along different loading paths• The methodology is now being implemented in two programmatic efforts: Multiscale Modeling and High Pressure Strength
Future Work• Based on this work a new technique has been proposed to investigate deformation properties under high strain rate, low pressure
loading conditions. If successful, the method could provide highly sought after data to constrain material constitutive models.
Publications“Shock-induced void collapse in fcc metals”, E.M. Bringa, et. al., Applied Physics Letters, In Press.“High pressure, laser driven deformation of an aluminum alloy,” J. M. McNaney, et. al., Met. Trans A, 35A, pp 2625-2631 (2004).
24 GPa
Shocklessloading
Shockloading
20 GPa8 GPa
20 GPa16 GPa8 GPa
24 GPa 31 GPa 42 GPa
Comparing calculations withexperimental results confirms the
validity of the simulations
A comparison of predicted versus measured crater depth ensures that theconstitutive models used are appropriate.
A novel geometry has been developed to assess the amount of meltingdue to the drive plasma-solid interaction. Comparison with simulationyields confidence in the conductivity models employed. This capabilityhas already been fielded elsewhere for programmatic efforts.
Recovered Sample
Focused ionbeam excision
Calculations indicate the amount ofmaterial that melts over the pressurerange of interest. This allows us toavoid regions affected by thermaltransients.
The transition from a shockless to shock loading path, which occurs atconstant pressure, is most easily seen in the accompanying temperaturerise. The transition takes place over a distance of about 50 µm
Recovered MeltSample (25 GPa)
Loading along the shock Hugoniotconstrains the pressure, temperature,strain, and strain rate. Accessing aloading path on or near the isentropeallows independent control of thepressure, temperature and strain rate.
Sample
Reservoir Evacuated gap(≈ 250 µm)
Laser(50-400J
The drive is generated by illuminating areservoir material using the Omega laser(Rochester, NY). Subsequently thereservoir material creates a smooth, shortpressure pulse in the sample to be studied.
Recovery Tube
Recovered Sample
Molecular dynamics provides insightinto the deformation mechanisms
During ramped loading, source activiation leads to lower dislocationdensity. Based on the mechanism above, this reduction is expected to leadto an increase in the pressure required for twin formation.
0.1 µm
Ramped versus Sharp Shock with Dislocation Sources (50 GPa)
Shock
0 psRise time
50 psRise time
Twin formation in shock driven molecular dynamics simulations has beenobserved. The mechanism is one of stacking fault coalescence, asproposed by Zaretsky [Appl. Phys. 78, 3740 (1995)].
Slices of the simulation box, perpendicular to the<110> direction, featuring <111> planes. Coloringproportional to centro symmetry parameter
t=10.8ps
001110
t=10.6ps
Nucleation of 3 adjacentstacking faults lead to 2twin boundaries
nano-twin
Removal by N. Telsich, LLNL
Simulation Overview
Radi
us (m
m)
2.0
1.5
1.0
0.5
0
-0.5
-1.0
-1.5
-2.0
10.0
8.0
6.0
4.0
2.0
0.0
Dens
ity (g
/cm
3 )
Axial position (mm)0.0 0.5 1.0 1.5
t = 88 ns
Deformation in Samples Loaded Isentropically
shock(z=120 µm)
shockless(z=40 µm)
Velocimetry measurements performed on thin foils yield informationon the pressure versus time profile and provide benchmarking forhydrodynamic simulations
Velocimetry Data(February 2005, shot 6)
Velocity and InferredPressure Profile
Frost-Ashby Deformation Map
Ti
The deformation mechanismsare the scientific underpinning ofthe constitutive models used todescribe material deformationbehavior
Dynamic Loading Paths inPressure-Temperature Space
Limits on length and time scales constrain MD simulations toloading paths along the the shock Hugoniot
Simulation of Material Response to 50 GPa ShockLoading
Simulated Melt Response
0
1
2
3
4
0 1 2 3Time (µs)
Mel
t Dep
th (µ
m)
Resolidificationbegins
measured
The experimental package is contained in a tube, facilitating materialrecovery. The sample size, large compared to the loading pulse width,ensures wave reflections are negligible.
Sample TubeCap
Incr
easi
ng v
eloc
ity
Increasing time
10 ns5.9
km/s
Comparison of Shock and Shockless Loading
Simulated Crater Formation
measured
Pressure Evolution Temperature Evolution
Loading Path Thermal History
25 GPa
20 GPa