#NXCAE13: University of California Riverside and Saratech

A Modern CAE Environment: Enabling Smarter Decisions

Planar Titanium Stent Design

Andrew Jabola, Application Engineer, Saratech Inc.

Shannon Gott (Ph. D. Candidate) & Masaru Rao (Assistant Professor), University of California - Riverside

NX CAE

Symposium

2013

Agenda

Objective

Background: Motivation for Stenting

Planar Stent Challenges

FEA: The Key to Redesign

Post-processing

Comparison Against Physical Test Data

Lessons Learned

2

Objective

Develop and refine titanium micromachining

techniques to create nanopatterned titanium

stents

Solve current stent limitations with physical means

Hypothesis: rationally-designed surface

nanopatterning will enhance desired vascular

cell responses relative to uncontrolled surfaces

3

Background: Heart Disease Heart disease is the leading cause of death in the U.S.

Most common form of heart disease is cardiovascular disease (CVD)

CVD is caused by atherosclerosis, characterized by plaque build up

Image from Texas Heart Institute: http://www.texasheart.org/hic/topics/cond/carotidarterydisease.cfm

4

Background: CVD Treatment

Three treatment options

Drugs

Catheter assisted procedures:

Stenting

Coronary artery bypass surgery

Image from National Heart Lung and Blood Institute:

http://www.nhlbi.nih.gov/health/dci/Diseases/stents/stents_all.html

5

Background: Stents

Today’s stents have a variety of problems

Stents damage artery wall lining

Potential complications

Bare-metal stents (BMS) : Restenosis – smooth muscle build up

Drug-eluting stents (DES) : Thrombosis – blood clotting

Images from Curfman GD et al., NEJM. 2007;256:1059-1060

6

Planar Stent Fabrication

Nanopatterning fabrication

techniques are inherently

planar

Using an approach

demonstrated by Takahata &

Gianchandani, we have

circumvented this limitation

7

Pattern Fabrication

Planar Stent Expansion

Planar to 3D stent

transformation

a) 80 µm thick, deep-etched

planar Ti stent

b) Tapered needle weaved

through to create compact

cylindrical geometry

c) Balloon-mounted stent

d) Stent after deployment in

3 mm I.D. mock artery

8

High uniformity of stent struts!

Planar Stent Radial Stiffness Testing

Radial stiffness is important for patency

Displacement control with load response

recorded

Current Ti stents possess about half the radial

stiffness of commercial stainless steel stents

Possible reasons for

lower radial stiffness:

Material Properties

Less material

Design 9

Stent Force-Displacement Data

FEA Objective

Analyze and Correlate current planar stent

design by Takahata against physical test data

Optimize future designs using FEA and verify

using physical test

10

Takahata Design Future Design

Analysis Challenges

Highly Nonlinear Analysis

Large Displacement/Large Strain

Difficult Contact Analysis both in Expansion and

Crushing

Nonlinear Material Properties

11

Analysis Objective

Expand Stent to match deployed

configuration

Stent is expanded to 3 mm in diameter

Crush Stent to correlate against physical test

data

Stent is crushed back to 1.5 mm for correlation

12

Analysis Setup

Pre/Post – NX 8.5 – Advanced Simulation

Solver – NX NASTRAN 8.5 Advanced

Nonlinear (ADINA) 601/129 NL Transient

Solution

Solution Timesteps (time/time steps)

Expansion (100 s/410 time steps)

Expander Relax (25 s/50 time steps)

Crush (75 s/200 time steps)

13

Analysis Setup

Material – Grade 1 Commercial Pure

Titanium

Isotropic, Plastic, Setup with a NL Stress-Strain

Curve

Stress-Strain Curve determined through tensile testing

by UC Riverside

Isotropic Hardening

14

Analysis Setup

Stress-Strain Curve

Data points were reduced from actual test data

15

0

50

100

150

200

250

300

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Str

ess (

MP

a)

Strain (mm/mm)

Titanium Stress-Strain Curve

Analysis Setup

Stent

Modeled using solid elements (~20K)

Reduced to a single link for reduction in

computation time

16

Analysis Setup

Needle/Balloon

Modeled using Plate Elements

Enforced Displacements used to expand

17

Analysis Setup

Crushing Mechanism

Modeled using plates and RBE2s

18

Analysis Setup

Elements

Overall Element size is 0.015 mm

Linear Solid (HEX/WEDGE) and Plate Elements

used

Plate Region is also the same resolution for

contact considerations

Element Count: 216732 elements

19

Analysis Setup

Boundary Conditions – Stent

Stent was constrained only in Y at ends

20

Y+

Analysis Setup

Boundary Conditions – Balloons

Enforced Displacements used on balloons

1) Balloons Initial Translate in Z directions to initially

expand stent

2) Center Plate Radial expands to fully deploy stent

21

Analysis Setup

Boundary Conditions – Initial Expansion

22

Analysis Setup

Boundary Conditions – Full Deploy (3 mm)

23

Analysis Setup

Boundary Conditions – Plates

Enforced Displacements at RBEs used for

crushing

24

1.5 mm

Analysis Setup

Boundary Conditions – Animation

25

Analysis Setup

Contact Conditions

Frictionless Contact

Contact Birth/Death Used for expansion and

crushing

26

Analysis

NX NASTRAN Adv. NL (ADINA) – 601/129

Transient Analysis

Solution Memory: 5.4 GB

27

Post-processing

Test Correlation Indicators

Visual Deformation

Crushing Reaction Force

28

Deformation

Analysis Results

Results

Deformation - Comparison

30

Results

Deformation - Comparison

31

Results

Deformation - Comparison

32

Results

Reaction Force – Crushing

Force Processed as N/mm of link length (link length

0.8 mm)

Reaction Force taken at end of RBE

33

Results

Reaction Force – FEA vs. Physical Test

34

Lessons Learned

Analysis was product of many runs

Following is list of important parameters and

lessons learned while running SOL 601/129

These can be applied to many other NL

analyses

35

Lessons Learned

General Tips

Step-by-Step – Analyze in pieces, don’t try and

setup entire analysis in one shot

Time Step Optimization – Place Time Steps only

where you need them

Automatic Time Stepping is a must (should be a

default)

36

Lessons Learned

Contact Parameters and Mesh Resolution

Line Search – Turn this on to help with contact

problems

Mesh Resolution – Meshes must match between

contact to get most accurate force results

Contact Damping is required, but needs to be

tuned, otherwise results can be odd

37

Future Work

Future Work is being carried on by Shannon

Gott of UC Riverside

Optimization of design to increase stiffness

Fatigue Life Analysis

38

Conclusion

Correlation between physical test data and

FEA was achieved using NX NASTRAN and

NX CAE

Many lessons learned that are applicable to

many nonlinear situations

Provides Basis for future work

39

Q/A

40

References 1. J. Lu, M. P. Rao, N. C. MacDonald, D. Khang and T. J. Webster,

Acta Biomaterialia, 2008, 4, 192-201.

2. P. Vandrangi, S. C. Gott, V. G. J. Rodgers and M. P. Rao,

presented in part at the 7th International Conference on

Microtechnologies in Medicine and Biology, Marina Del Rey, CA,

April 10 – 12, 2013, 2013.

3. A. W. Martinez and E. L. Chaikof, Wiley Interdiscip. Rev.-Nanomed.

Nanobiotechnol., 2011, 3, 256-268.

4. M. F. Aimi, M. P. Rao, N. C. Macdonald, A. S. Zuruzi and D. P.

Bothman, Nat. Mater., 2004, 3, 103-105.

5. E. R. Parker, B. J. Thibeault, M. F. Aimi, M. P. Rao and N. C.

MacDonald, J. Electrochem. Soc., 2005, 152, C675-C683.

6. K. Takahata and Y. B. Gianchandani, J. Microelectromech. Syst.,

2004, 13, 933-939.

41