Characterization of Additively Manufactured Ti-6Al-4V for ... آ  Characterization of Additively

  • View
    0

  • Download
    0

Embed Size (px)

Text of Characterization of Additively Manufactured Ti-6Al-4V for ... آ  Characterization of Additively

  • Characterization of Additively Manufactured Ti-6Al-4V for Use in Pacemaker Shields Alexander Antonuccio, Charlie Meisel, Fiona O’Dowd, Alyssa Stubbers Faculty Advisors: Dr. David F. Bahr, Dr. Ernesto E. Marinero Industrial Sponsors: Jordan Balmer (Medtronic), Dr. Peter Tortorici (Medtronic) & Dr. John Barnes (The Barnes Group) Acknowledgements: David Brice (Purdue), PulseTech Products Corp.

    Medtronic evaluated two suppliers to determine if their additive manufacturing process was a suitable replacement for the current method of deep drawing custom pacemaker shields. A heat treatment softened the as-printed fine, brittle microstructure. After heat treatment, 3D Systems shields had a point defect density comparable to the deep drawn samples. After heat treatment, TransMachine shields had a microhardness most like the deep drawn samples. 3D Systems pores had a lower aspect ratio and average pore size (27.9um), with a more predictable pore formation and geometry compared to TransMachine (49.1um).

    MSE 430-440: Materials Processing and Design

    Shields from two suppliers, 3D Systems (3DS) and TransMachine (TM), were examined.

    This work is sponsored by Medtronic plc, (Mounds View, MN) and The Barnes Group (Pittsburgh, PA)

    Titanium (Ti-6Al-4V) shields provide a barrier between the body and internal pacemaker electronics

    1) Characterize shields from suppliers to determine similarities to deep drawn

    2) Develop heat treatment for microstructure similar to deep drawn

    3) Advise on concerns of porosity, brittleness, and residual stress

    Current Custom/Prototype: Machine from Solid Block

    Future Custom/Prototype: Additive Manufacturing

    Custom shapes are required for prototyping and select patients

    o Cost ↓ 25%

    Heat Treatment Rapid cooling characteristic to additive manufacturing leads to fine, brittle microstructure Annealing and cooling slowly results in grain growth

    Sample Heat Treatment: Hold 3 hours at 1100°C Cool at 5°C/min

    X-Ray Diffraction (XRD) Diffraction scans on all samples elucidated microstructure

    Full Width at Half Max (FWHM) analysis gives information on point defects

    Microscopy SEM - Lineal analysis on BSE micrographs of shield cross sections

    Optical – Pore size analysis on shield cross sections Knoop Microhardness

    Long aspect ratio of Knoop tip appropriate for thin shields

    Hardness scales with strength and will indicate brittleness

    Project Background Results & Discussion

    Objectives

    Procedures

    Results & Discussion (cont.)

    Recommendations

    Heat Treatment Softening

    Porosity Analysis

    Conclusions

    Higher FWHM =

    Higher density of point defects

    = Lower ductility

    Heat treatment reduced the FWHM value for both suppliers. As-deposited, 3DS shields had a higher FWHM than TM. After heat treatment, 3DS shields had a lower FWHM than TM and were closest to the deep drawn shields. Shield thickness and location of scan had no effect. 3DS shields with additional processing steps have higher FWHM values, likely leading to more brittle shields.

    3DS TM standard 0.016” standard 0.020” HIP’ed 0.020” standard 0.012” stress relief 0.020”

    Characterization

    Shields as deposited had a fine lamellar α + β microstructure.

    XRD peaks confirmed the above microstructure. No other phases were present.

    TM shields consistently had a finer microstructure than 3DS shields. Heat treatment increased grain size significantly in all parts, decreasing the grain boundary strengthening effect. This change was greater for 3DS.

    After heat treatment, only TM shields are statistically softer than the required 365 HK specification; the 3DS shields were not. Location and shield thickness did not effect microhardness.

    Wide distribution of pore sizes in TM shields make porosity harder to predict.

    Sharp pore morphology is more likely to lead to crack propagation and part failure. 3DS pores had a lower aspect ratio, reducing likelihood of fracture initiation. Pores did not form preferentially in certain areas of the shields of either supplier.

    Residual stress measurements on two regions of a 3DS shield resulted in measurements of -599 MPa and -668 MPa. Both values are more than half the yield strength of Ti-6Al-4V.

    Residual Stress

    o XRD and SEM confirmed additively manufactured shields had a lamellar α + β microstructure.

    o While heat treatment revealed point defects for both suppliers, the heat treated 3DS shields were most similar to deep drawn shields.

    o Heat treatment increased grain size which decreases the Hall-Petch effect on hardness, leading to more ductile shields.

    o The Knoop microhardness data agreed with the SEM and XRD data.

    o Heat treated TM shields were the only samples below the maximum microhardness of 365 HK

    o 3DS pores are less likely to initiate crack propagation due to their equiaxed morphology

    o TransMachine pores are large and unpredictable, making bridging a concern

    o As-deposited shields have high residual stress o As delivered from suppliers, shields do not

    have desired properties for use in pacemakers o The designed heat treatment resulted in more

    favorable shields that were not all within specification

    3D Systems • Heat treatment to soften and reduce residual

    stress TransMachine • Modification of printing parameters to reduce

    porosity While the designed heat treatment was effective, a shorter process should be researched

    Current Production: Deep Drawing

    Medtronic plc, “Our Pacemakers”

    fast cool = martensite

    slow cool = α + β

    TM3DS

    as -d

    ep os

    ite d

    he at

    tr ea

    te d

    T. Ahmed, H.J. Rack. Materials Science and Engineering A. (243) 1998

    █ TM (n=197 ; x̄ = 49.1 μm) █ 3DS (n=98 ; x̄ = 27.9 μm)

    25 30 35 40 45 50 55 60 65 70 0.0

    0.5

    1.0

    In te

    ns ity

    (N or

    m al

    iz ed

    )

    Deep Drawn Shield

    o Lead time 4-6 weeks → 24 hours

    20 25 30 35 40 45 50 55 60 65 70 75 80 85 0.0

    0.5

    1.0

    In te

    ns ity

    (N or

    m al

    ize d)

    TransMachine .020" Back α Titanium Phase Peaks

    Dee p D

    raw n b

    ase

    3DS ba

    se

    3DS HT

    ba se

    TM 0.0

    20" Fro

    nt B ase

    TM 0.0

    20" Ba

    ck B ase

    TM 0.0

    12" Fro

    nt B ase

    TM 0.0

    12" Ba

    ck B ase

    TM 0.0

    20" Ba

    ck B ase

    HT

    TM 0.0

    12" Fro

    nt B ase

    HT

    TM 0.0

    12" Ba

    ck B ase

    HT

    3DS SP

    Ba se

    Fro nt

    3DS SP

    Ba se

    Bac k

    3DS HIP

    Ba se

    Fro nt

    3DS HIP

    Ba se

    Bac k

    0.0

    0.5

    1.0

    1.5

    0.14

    0.93

    0.29

    0.73

    0.55

    0.75 0.74

    0.48 0.54 0.51

    1.17

    1.01

    1.13 1.19

    3DS Post ProcessTM HTTM

    M ea

    n FW

    H M

    HT = Heat Treated TM = TransMachine 3DS = 3D Systems

    3DS

    3D S .

    01 6"

    cou po

    n

    3D S .

    01 6"

    ba se

    3D S .

    01 6"

    HT

    TM A .

    02 0"

    ba se

    TM A .

    02 0"

    HT

    TM A .

    01 2"

    ba se

    TM A .

    01 2"

    HT

    0

    5

    10

    15

    20

    25

    30

    Di st

    an ce

    b et

    we en

    β g

    ra in

    s (µ

    m )

    3DS TM

    De ep

    D raw

    n

    3D S

    St d.

    3D S

    HT

    3D S

    We ld

    We ld

    Co up

    on

    TM .0

    12 in

    TM .0

    20 in

    TM H

    T

    0

    100

    200

    300

    400

    500

    M ic

    ro ha

    rd ne

    ss (H

    K )

    3DS TM

    dark phase: α

    light phase: β

    Instron, “Instron Knoop Test”

    Slide Number 1