National Laboratory - Krell Inst

Lawrence Livermore

National Laboratory 1

Joshua Kuntz, Eric Duoss, Wayne King, Melissa Marggraff,

Robert Maxwell, Tom Wilson, Matt Wraith, Todd

Weisgraber, Andrew Pascall, Cheng Zhu, Rayne Zheng,

James Frank, Joshua DeOtte, Chris Harvey, Tom Metz,

Kyle Lange, Marcus Worsley, George Farquar, Tammy

Olson, Sergei Kucheyev, Chris Orme, Kyle Sullivan, William

Smith, Maxim Shusteff, Luis Zepeda-Ruiz, Scott Fisher,

Tom Wilson, Alex Gash, & John Vericella

Academic Partners: Professors J. Lewis (Harvard), N. Fang

(MIT), D. Tortorelli (UIUC), J. Hopkins (UCLA)

LLNL-PRES-646626

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Lawrence Livermore National Security, LLC

Lawrence Livermore


Additive Manufacturing

Commercial AM tools, custom AM processes

Synthesis

Tunable nanomaterials, nanoparticles, crystal

growth, polymers, aerogels, feedstocks

Modeling and Design

HPC, multi-scale, multi-physics, topology

optimization, analytical design

Predator UAV, Air Force

Certification

Process modeling, in-situ characterization

By integrating optimization, design and modeling, tailored synthesis, and additive

manufacturing methods, we can enable high performance materials and components.

Lawrence Livermore


Order-of-magnitude performance improvement and decoupling of material properties is possible

102

1

10–2

10–4

Yo

un

g’s

mo

du

lus

(sti

ffn

ess

), G

Pa

1 10 102 103

Density, kg/m3

Designed microarchitecture Stretch-dominated lattices can provide these properties

Lawrence Livermore


Material selection chart for thermal expansion and stiffness

Jefferson, et al., Int.

Jrl. Of Solids and

Structures, 2009.

Steeves, et al., Jrl. Of Mechanics and

Physics of Solids, 2007.

Lakes, et al.,

APL, 2007.

Theoretical

structures:

Lawrence Livermore


1

10

100

1000

10000

0 2 4 6 8

Pro

pagation v

elo

city (

m/s

)

Energy density (kcal/cc)

Monomolecular

materials

Binary materials and thermites

Al/KClO4

HMX

PETN

TNT

Al/CuO Al/MoO3

Desired

performance

Co

ntr

ol

of

str

uctu

re

Reduce particle size and

optimize organization of

fuel/oxidizer microassembly

Irregular distribution of binary

constituents limits performance

Al

2 µm

Highly ordered and optimized

microstructure

MoO3

Al

Co

ntr

ol

of

str

uctu

re

10’s µm

Lawrence Livermore


Lattice of unit cells

Topology optimization A computational design method

Freedom and constraint topologies

An analytical design method

Multimaterial microlattices with prescribed thermal expansion coefficient

Al

ABS

Void

Unit Cell Periodic Material

LLNL is developing new manufacturing technologies and using HPC to

deterministically design and fabricate architected materials.

Lawrence Livermore

National Laboratory 7 A broad range of extreme thermo-structural properties can be rapidly designed.

Hopkins, et al., “Polytope Sector-based Synthesis and Analytical

Optimization of Microstructural Architectures with Tunable Thermal

Expansion,” manuscript under review.

Lawrence Livermore


Material is comprised of unit cells arranged periodically.

Mesh a unit cell

• Assign each element phase concentrations

• Interpolate to obtain element properties

Use homogenization theory to determine unit cell properties

Use optimization algorithm to assign the concentrations to

• Minimize the cost function (e.g. RMS error)

• Satisfy constraints (e.g. isotropy)

Unit Cell Periodic Material Finite Element

Lawrence Livermore


Aluminum

Invar

Void Constraints:

• Minimize CTE subject to lower bound on

stiffness

• Maintain vol. fraction ~25% for solids

CTE value

Lawrence Livermore


• Copper/ PMMA system

• CTE = -1.3e-6 mstrain/K

• Thermal conductivity = 57 W/m-K

• Density of 2600 kg/m3

• Stiffness = 21.3 GPa

• Manufacturing constraints are included

• Copper phase is continuous due to thermal conductivity constraint

LLNL’s computational resources make 3D designs and additional

physical constraints possible.

Topology optimization is being used to design microlattice-based materials

for specific applications requiring additional physics and constraints.

Lawrence Livermore


Unit cell architecture at various heights

3D unit cell with refinement

We are advancing the state of the art in topology optimization beyond currently

available capabilities by utilizing HPC.

Actual fabricated component

(2 polymers)

Lawrence Livermore


Projection Microstereolithography (PµSL) A photochemical and optical technique

Direct Ink Writing (DIW) Utilizes unique flow and

gelling properties

Electrophoretic Deposition (EPD) Electric fields transport nanoparticles

LLNL has been developing a combination of additive micro- and nanomanufacturing technologies

Lawrence Livermore


Projection Microstereolithography (PµSL) A photochemical and optical technique

Direct Ink Writing (DIW) Utilizes unique flow and

gelling properties

Electrophoretic Deposition (EPD) Electric fields transport nanoparticles

LLNL has been developing a combination of additive micro- and nanomanufacturing technologies

200 mm

200 µm

5 mm 5 mm

Lawrence Livermore


Projection Microstereolithography (PµSL) - a photochemical and optical technique

Lawrence Livermore


Zheng, et al., “Ultralight, ultrastiff mechanical

metamaterials,” Science, June 20, 2014.

Lawrence Livermore


Solid polymer

r ~ 80 kg/m3

(11% rel. density)

Hollow tube Ni-P

r ~ 40 kg/m3

(0.5% rel. density)

Hollow tube Al2O3 (ALD)

r ~ 0.9 kg/m3

(0.025% rel. density)

Solid (sintered) Al2O3

r ~ 320 kg/m3

(8% rel. density)

Zheng, et al., “Ultralight, ultrastiff mechanical metamaterials,” Science, June 20, 2014.

Lawrence Livermore


This architected material has orders of magnitude higher stiffness than any other

material in this low density regime.


Lawrence Livermore


Heterogeneous multi-material micro-structures

500μm 1 mm

Designed

heterogeneous helix

HD

DA

Optical image of

fabricated part

diameter = 1 mm

Encapsulation of functional particles

11% octet truss

2mm 2mm

20% octet truss

3 mm

3D multimaterial Al2O3

polymer

Negative CTE

multimaterial

architectures

designed using

our analytical

method.

3D multimaterial microarchitectures are enabled by our unique additive

manufacturing platform.

Lawrence Livermore


Direct Ink Writing (DIW) - utilizes unique flow and gelling properties

Lawrence Livermore


Direct write of a siloxane material with designed porosity and structure for specific compressive and shear properties –

control of filament size, pitch, and microarchitecture

Large square front zoom

(scale in 100ths of an inch)

Simple cubic

Face centered tetragonal

Soft materials with tailored mechanical response have broad application for NNSA.

Lawrence Livermore


“Simple Cubic” Axial Response “Centered Tetragonal” Axial Response

Duoss, et al., “Three dimensional printing of elastomeric cellular architectures with negative

stiffness,” Advanced Functional Materials, published online April, 2014.

1 mm 1 mm

Lawrence Livermore


Isostress contours reveal columnar stack up of stress in simple cubic

geometry whereas stress is more distributed in face centered tetragonal.

Modeling and simulation has reveal a possible instability in the simple cubic architecture.



Lawrence Livermore


“Centered Tetragonal”

“Simple

Cubic”

Negative stiffness region

“snap through” effect

Simple cubic

Face centered tetragonal



Lawrence Livermore


1 cm 1 cm

1 cm

Dimensionality: DIW is now capable of

patterning arbitrarily complex

structures by co-printing a fugitive

support material that is removed after

printing

Material Set: Al, Al2O3, CuO, B4C,

Bi2Te3, Sb2Te3 and silicone-based inks

have been developed.

Octet-truss

unit cell

Inter-digitated

structure

“Inter-woven”

lattice

Lawrence Livermore


Electrophoretic Deposition (EPD) - electric fields transport nanoparticles

Lawrence Livermore


Drop-cast versus EPD Al-CuO thermite

• EPD samples exhibit ~2 the power density of their drop-cast counterparts (~½ burn time and ~2 burn velocity)

• Flame from EPD Al-CuO thermite propagates around progressively tighter corners

Bend test for EPD Al-CuO thermite

10 mm

10 mm

Drop-cast film

EPD film 0 ms 5 ms 10 ms 15 ms

EPD

0 ms 5 ms 10 ms 15 ms

Drop cast

~5 cm

~5 cm

Lawrence Livermore


Channels

Hurdles

“Hurdles” v = 42 m/s

“Channels” v = 138 m/s

High-speed optical imagery Thermal imagery

1 mm

1 mm

Lawrence Livermore


Small A

Large A

L = ½ film thickness

Data shows that there is more than one relevant length-scale.

K. T. Sullivan, J. D. Kuntz and A. E. Gash, JAP, 2013.

Lawrence Livermore


1

10

100

1000

10000

0 1 2 3 4 5 6 7 8

Pro

pagation V

elo

city (

m/s

)

Energy Density (kcal/cc)

MONOMOLECULAR MATERIALS

HMX

PETN

TNT

DESIRED

PERFORMANCE

Micron-Al + nano-CuO EPDed

Nano-Al + nano-CuO EPDed

Nano-Al + nano-CuO EPDed on DIW wide strips

Nano-Al + nano-CuO EPDed in DIW channels

Nano-Al + nano-CuO EPDed and confined in microcapillary

Al / CuO

Nano-Al + nano-CuO EPDed and confined in microcapillary

Lawrence Livermore


LLNL research and

development directions:

• Microarchitecture and

“New” Material

Properties – “Boutique

Materials”

• Next Generation

Processes

• Qualification &

Certification

• Exotic materials

• Process Understanding

& Performance Modeling

National Laboratory - Krell Inst

Documents

Transcript of National Laboratory - Krell Inst