15 Medical Voroni Niti

Subscriber access provided by READING UNIV
Article
Design and Compressive Behavior of Controllable Irregular Porous
Scaffolds: Based on Voronoi-Tessellation and for Additive Manufacturing
Guanjun Wang, Lida Shen, Jianfeng Zhao, Huixin Liang, Deqiao Xie, Zongjun Tian, and Changjiang Wang
ACS Biomater. Sci. Eng., Just Accepted Manuscript • DOI: 10.1021/acsbiomaterials.7b00916 • Publication Date (Web): 16 Jan 2018
Downloaded from http://pubs.acs.org on January 18, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Biomaterials Science & Engineering is published by the American Chemical

Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 28 ACS Biomaterials Science & Engineering
1
2
3
4
5
6
7
8
Design and Compressive Behavior of Controllable
9
10
11
12
Irregular Porous Scaffolds: Based on Voronoi-
13
14
15
16
Tessellation and for Additive Manufacturing
17
18
19
20
Guanjun Wang1,2, , Lida Shen1, , Jianfeng Zhao1,*, Huixin Liang1, Deqiao Xie1, Zongjun Tian1,
‡ ‡
21
22
23 and Changjiang Wang3
24
25
26 1
College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and
27
28
29 Astronautics, 29 Yudao Street, Nanjing 210016, PR China.
30
31
32 Suzhou Kangli Orthopedics Instrument Co.Ltd, Luyuan Tangqiao Town, Zhangjiagang
2
33
34 Suzhou, 215600, PR China.
35
36
37
38
3
Department of Engineering and Design, University of Sussex, Sussex House, United Kingdom.
39
40
41 E-mail: zhaojf@nuaa.edu.cn
*
42
43
44 KEYWORDS: Porous scaffold, Voronoi-Tessellation, Selective Laser Melting, Stress shielding,
45
46
47
Elastic modulus, Gradient pore structure.
48
49
50 ABSTRACT: Adjustment of the mechanical properties (apparent elastic modulus and
51
52
53 compressive strength) in porous scaffolds is important for artificial implants and bone tissue
54
55 engineering. In this study, a top-down design method based on Voronoi-Tessellation was
56
57
58
59 1
60 ACS Paragon Plus Environment
ACS Biomaterials Science & Engineering Page 2 of 28
1
2
3 proposed. This method was successful in obtaining the porous structures with specified and
4
5
6 functionally graded porosity. The porous specimens were prepared by selective laser melting
7
8 technology. Quasi-static compressive tests were conducted as well. The experiment results
9
10 revealed that the mechanical properties were affected by both porosity and irregularity. The
11
12
13
irregularity coefficient proposed in this study can achieve good accommodation and balance of
14
15 ‘irregularity’ and ‘controllability’. The method proposed in this study provides an efficient
16
17 approach for the bionic design and topological optimization of scaffolds.
18
19
20
21 1. Introduction
22
23
24
Long-term aseptic loosening of artificial implants and refracture are fatal complications within
25
26
27 the clinical setting1-2. The main issue within artificial implants is ‘stress shielding’. Stress is
28
29 unable to adequately transfer from the metal implant device to the bone. This is caused by the
30
31 large difference between their elastic moduli. Osteocytes that do not have adequate stress
32
33
34 stimulation will die and be absorbed, which results in the loosening of implants and sometimes
35
36 fracturing occurs. Porous structures were introduced to the implant devices to stop these
37
38 occurrences. These structures could reduce the apparent elastic modulus to the level of human
39
40
41
bones (1 Gpa-10 Gpa)2. The porous structures also provide necessary space for osteocyte and
42
43 transport pathways of tissue fluid. This is conductive to biological fixation between the bone and
44
45 implant3-5. Metal implants with porous structures could be fabricated rapidly and precisely as
46
47
additive manufacturing (AM) has become more developed. The selective laser melting (SLM)
48
49
50 and the electron beam melting (EBM) are the most popular methods within AM. Designing
51
52 porous scaffolds that could meet both structural and mechanical properties requirements has
53
54 become necessary6.
55
56
57
58
59 2
1
2
3 The morphology of these porous structures are broadly divided into regular or irregular
4
5
6 structures. The primary construction methods of the regular porous structures are the cell unit
7
8 method7 and the triply periodic minimal surface method8. The regular porous scaffolds have
9
10 regular pore morphology, good connectivity, and controllable mechanical properties. This results
11
12
13
in their wide application9. Irregular porous scaffolds are typically implemented by computer
14
15 programs and mathematical models10-12. The irregular porous scaffolds enable geometric and
16
17 mechanical parameters to be designed precisely or distributed in a gradient manner. Furthermore,
18
19
irregular porous scaffolds have the ability to simulate the complex and anisotropic
20
21
22 microstructures of bone tissues. This provides additional freedom within bionic design and
23
24 topological optimization13-15. The design approach for irregular porous structure is another key
25
26 technology that requires additional research.
27
28
29
30 Unfortunately, most of previous studies on irregular porous structures ignored the effect of
31
32 irregularity and its evolution. Uncontrollable irregularity leads to poor repeatability of
33
34 morphology and the mechanical properties. In this study, we proposed a top-down design
35
36
37
method (probability sphere method) based on the Voronoi-Tessellation method to construct
38
39 controllable porous scaffolds. The probability sphere method used a regular three-dimensional
40
41 point array that was agitated by probability spheres. This generated the irregular but controllable
42
43
porous scaffold. In order to evaluate the porous scaffolds, we prepared two sets of specimens by
44
45
46 SLM. We then measured their geometric parameters with industrial CT (Computer Tomography).
47
48 Their mechanical parameters were tested with quasi-static compressive experiments.
49
50
51
52
2. Materials and Methods
53
54
55 2.1 Modeling and characterization of porous scaffolds
56
57
58
59 3
1
2
3 The Voronoi-Tessellation is a method of space partition, based on the seed point16-18. The
4
5
6 Voronoi diagram is defined below:
7
8 For a set of points on the m-dimensional Euclidean space
9
10 m
11
P  { p 1 , , p n }  R , 2  n  , p i  p j ,i , j  I n  {1, , n} (1)
12
13
14 The line between pi and pj divides the space into two parts, Hi(pipj) is the part including pi, then,
15
16
17
18 V ( p i )  {x x  p i  x  p j }   H(p i, p j) (2)
jI n \{i}
19
20
21
22 is the m-dimensional Voronoi generated by pi on Rm space , γ(p) = {V (pi),..., V(pn)} is called
23
24 the Voronoi diagram and pi is the seed point. Voronoi diagram is determined by the amount and
25
26 the distribution of seed points. The control of the seed points is paramount to successful irregular
27
28
29 porous scaffolds modeling. Previous researches have preferred to generate three dimensional
30
31 points in random manner19-21. However, there is little control of these random points and the
32
33 morphology and the mechanical properties have poor repeatability.
34
35
36
37
In this study, the controllable irregular porous structures were designed using CAD software
38
39 Grasshopper version 0.9.0076 (http://www.grasshopper3d.com). First, a cube region was
40
41 generated and the volume of the cube region was the outer volume of the porous structure.
42
43
Second, an ordered cube lattice containing n layers was generated within the cube region. The
44
45
46 distance between the two adjacent points within Layer i was bi, as seen in Figure 1. Denote the
47
48 point on Line m and Column n in Layer i as Pmn_i. Denote the space between Layer (i-1) and
49
50 Layeri as ai. The spherical regions that were centered at each point and generated with the radius
51
52
53 of Ri. Third, a new point P’mn_i was then generated randomly in each spherical region to replace
54
55 their predecessors. This was the new lattice. P’mn_i was generated as follows:
56
57
58
59 4
1
2
3
4  x  x  R  rand  sin(  rand )  cos(2  rand )

5  y  y  R  rand  sin(  rand )  cos(2  rand ) (3)
6  z   z  R  rand  sin(  rand )  cos(2  rand )
7 
8
9
10 Where, rand is a random function with uniform distribution in [0, 1].
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 Figure 1. Regular lattice. (a) Perspective view. (b) Front view. (c) Top view of Layer i.
27
28
29 Irregularity coefficient was defined as ε:
30
31
32 
33 ，
34 Pmn_iP mn_i
1
35
36
 
N mni ai
37 ,, (4)
38
39
40 Where, N is the number of points in the irregular lattice generated previously, 0 ≤ ε < 1.
41
42
43
44 The radius of spherical region Ri was constrained to increase the controllability of the irregular
45
46 1
47
lattice by 0  Ri  a . The Voronoi cells were generated in Grasshopper following the
2 i
48
49
50 irregular lattice generation. The struts were generated based on the edges of the Voronoi cells.
51
52 Then, a Boolean operation was used to form a porous scaffold with a specific shape, as shown in
53
54 Figure 2. The size of the pores and struts were controlled by introducing the scale coefficient K.
55
56
57
58
59 5
1
2
3
4 This was the ratio of a pore area to the corresponding surface area of the cell K  S ci , as
S pi
5
6
7 shown in Figure 3. K controls both diameters of the pores and the struts. When the seed points
8
9 are given, the diameters of the pores and struts could be adjusted by K.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32 Figure 2. The design principle of porous scaffold. (a)Designing regular lattice. (b)Generating
33
34 probability spheres. (c)Obtaining irregular lattice. (d)Generating voronoi cells based on irregular
35
36
37 lattice. (e)Constructing porous structures.
38
39
40
41
42
43
44
45
46
47
48
49
50 Figure 3. The definition of scale coefficient K. (a) One of the Voronoi cell. (b) Porous
51
52 structure based on (a).
53
54
55
56
57
58
59 6
1
2
3 The gradient distribution of the points in the Z axis were obtained. This allowed one to obtain the
4
5
6 gradient distribution of porosity in the direction of Z axis, as shown in Figure 4. K was then
7
8 replaced with linear variables that related to cell position:
9
10
11 Zt  Zb Z  Zb
12 K  Kb  t  Zb (5)
Kt  K b Kt  K b
13
14
15
16 Where, Zb and Zt denoted the minimum and the maximum of the Z axis coordinate of body
17
18 center of all the voronoi cells. Kb and Kt represented the minimum and maximum of the value of
19
20
21
K. In order to fit the gradient distribution of points, the radius of probability sphere was set as
22
23 1
Ri  a . Based on the probability sphere method, Figure 5 shows the process of constructing
24 2 i
25
26
27
the artificial bone with porous gradient structures.
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Figure 4. The modeling method of porous scaffold with gradient porosity. (a) The regular lattice.
51
52 (b) The irregular lattices. (c) Porous scaffolds with gradient distribution of porosity (Kt = 0.9, Kb
53
54
= 0.7).
55
56
57
58
59 7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15 Figure 5. The method of constructing the artificial bone with porous gradient structures: Firstly,
16
17
18
offset the surface by incremental distances; Secondly, generate spherical regions based on the
19
20 intersection points of UV isocurves; Thirdly, generate irregular lattice then construct porous
21
22 gradient structures.
23
24
25
The parameters of the porous scaffold included pore size, porosity, and strut diameter. The
26
27
28  V 
porosity in the geometric model was calculated by the formula   1  P   100% , where Φ is
29 V  
30
31
32
porosity, VP is the volume of the porous scaffold, and V is outer volume of the porous scaffold.
33
34
35 The sections at the midpoint of the strut sections were extracted. The equivalent diameter method
36
37 was used to calculate the pore diameters when the area was equivalent to the sections. The
38
39
equivalent diameter method was used to calculate the aperture, as shown in formula(6) .
40
41
42
43  1 n 4Si
D= n  
44  D i 1
45  n 4A
(6)
d = 1
46 
 n i 1 
i
47  d
48
49
50 Where, nD is the number of pores, nd is the number of struts, D is the equivalent aperture, Si is
51
52
the hole area, d is the equivalent strut diameter, and Ai is the cross-sectional area of strut.
53
54
55
56
57
58
59 8
1
2
3 2.2 Fabrication and compressive testing
4
5
6 The porous specimens were fabricated with an SLM machine (M290, EOS GmbH, Germany)
7
8 using the optimized processing parameters, as shown in Table 1. The materials were commercial
9
10 Ti6Al4V ELI powders, supplied by EOS GmbH.
11
12
13
14
Table 1. SLM process parameters
15
16
17 Laser Scanning Hatch Layer
18 Laser focus
Parameter power speed Spacing thickness Atmosphere
19
20 /mm
/W / mm/s /mm /mm
21
22
23 Value 180 1350 0.1 0.03 0.08 Ar
24
25
26
27
28 The factors that influenced the mechanical properties of the porous scaffolds included porosity,
29
30
31
pore diameter, strut diameter, and shape of the strut section. Porosity is thought to be the most
32
33 influential factor22. Irregularity is another important parameter within irregular porous scaffolds.
34
35 In this paper, two sets of experiments were conducted to investigate the effects of porosity and
36
37
irregularity on the mechanical properties of porous scaffolds. The design parameters of the
38
39
40 porous scaffold are shown in Table 2. Three identical samples were prepared for each model and
41
42 the test results were average values of three samples.
43
44
45 Table 2. Design parameters of porous scaffolds
46
47
48
49 Group N K R
50
51 0.65 1
52 Porosity
53 125 0.7 1
54 series
55 0.75 1
56
57
58
59 9
1
2
3
0.8 1
4
5
0.85 1
6
7
0.9 1
8
9
10
0.8 0.4
11
12
0.8 0.8
13 Irregularit
0.8 1.2
14 y 125
15
0.8 1.4
16 series
17
0.8 1.6
18
19
0.8 1.8
20
21
22
23
24
25 X-ray computed tomography (XCT) is a effective nondestructive testing method in the field of
26
27 AM. Resolutions from 9.68 µm to 15.34 µm were performed by an industrial CT (XTH225,
28
29
Nikon, Japan) to scan the porous scaffolds. A 3D model of the porous scaffolds was
30
31
32 reconstructed to measure the porosity in software, as shown in Figure 6.
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Figure 6. Reconstruction of 3D model by XCT.
55
56
57
58
59 10
1
2
3 The compression tests of specimens were conducted with a mechanical testing machine
4
5
6 (CMT5105, MTS System Corporation, America). The crosshead displacement speed was fixed at
7
8 0.25 mm/min.The apparent elastic modulus (E) of porous specimens was characterized by quasi-
9
10 elastic gradient (ISO13314:2011), and the compressive strength (S) was characterized by
11
12
13
ultimate compressive stress, as shown in Figure7.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32 Figure 7. Characterization of stiffness and compressive strength.
33
34
35 In order to fit scattered data of mechanical properties, we used the Gibson-Ashby model22. The
36
37 Gibson-Ashby model is a typical model that simplified the irregular porous structure into the
38
39
40
regular structure built by the stacking of a special basic unit. This model discovered the
41
42 following functions between porosity, elastic modulus, and compressive strength:
43
44
45  Φ m
46  S  k1 (1  100 ) S 0
47  (7)
48  E  k (1  Φ ) n E
 2
100 0
49
50
51
52 Where, S0 and E0 is the compressive strength and elastic modulus of substrate materials, k1 and
53
54 k2 are constants related to mechanical properties of substrate materials.
55
56
57
58
59 11
1
2
3
4
3. Results and discussion
5
6
7 3.1 Controllability of geometric parameters
8
9
10 Figure 8 shows lattices and corresponding porous scaffolds with different regularities. Figure 9
11
12
13 depicts the relationship between the irregularity (ε) and (R=a) of probability spheres. The porous
14
15 scaffolds increase in irregularity as irregularity grows. There was a good linear relationship
16
17 between R=a and irregularity. This suggests that the probability sphere method could control and
18
19
20
characterize the degree of irregularity of porous scaffold. Figure 10 shows the influence of R/a
21
22 on porosity. When R/a is more than 0.2, porosity stay stable with the increase of R/a.
23
24 In theory, the irregular coefficient ε defined in this paper does not fully characterize the true
25
26 
27 degree of irregularity of the points. For example, if all vectors pi p’i were unidirectional,
28
29
30 although the value of ε could be huge, the degree of irregularity of the lattice would not change.
31
32 However, there is little prospect of such special case occurring in practical experiments. As a
33
34
35
result, the irregular coefficient ε meets the application requirements.
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 12
1
2
3 Figure 8. Porous structures with different irregularities. (a) ε = 0.06. (b) ε = 0.25. (c) ε =0.43. (K
4
5
6 = 0.9, N = 512).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25 Figure 9. Relationship between R=a and ε.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44 Figure 10. Relationship between R/a and porosity.
45
46
47
48 Figures 11 and 12 show the effect of the amount of the seed points (N) and scale coefficient (K)
49
50 on geometric parameters. seed points had little effect on porosity. The overall fluctuation range
51
52 was less than 10%. The larger seed point number was, the less the effect of it had on porosity. A
53
54
55
power relation existed between seed points and diameters of pores and struts. There were
56
57
58
59 13
1
2
3 strongly linear relation between scale coefficient and porosity, as well as diameters of pores and
4
5
6 struts. Scale coefficient had greater effect on geometric parameters than seed points. The primary
7
8 factor that affected the geometric parameters of porous scaffolds was scale coefficient.The
9
10 standard for the proper size of bone cell ingrowth is 200 µm to 1200 µm23. The method presented
11
12
13
in this paper adjusted seed points, scale coefficient and R=a to be within this scope.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Figure 11. The relation between points and (a) porosities, and (b) diameters of pores and struts.
32
33 (K = 0.8, R = 0.5).
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Figure 12. The relation between scale coefficients and (a) porosities, and (b) diameters of pores
53
54 and struts. (N = 216, R = 0.5).
55
56
57
58
59 14
1
2
3 3.2 Controllability of gradient porosity
4
5
6
7 Table 3 shows the four groups of Kb and Kt (see Figure 4). These groups were set to explore
8
9 the influence of different combinations on the gradient change of porosity. The slope of the best
10
11 fit line was defined as the porosity gradient G. Figure 13 shows the gradient variation of the
12
13
14
porosity as it related to height. As seen in Figure 10, the porosity changed linearly, in the height
15
16 direction which corresponded with K. When Kt remained constant, the maximum porosity
17
18 remained constant as well (the porosity at the top of the structure). This further confirmed that
19
20
porosity was primarily determined by K in 3.1.
21
22
23 Table 3. Scale coefficients for gradient porosity
24
25
26 Series Kt Kb
27
28 Ⅰ 0.9 0.6
29
30
Ⅱ 0.9 0.7
31
32
33 Ⅲ 0.9 0.8
34
35 Ⅳ 0.9 0.9
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56 Figure 13. Porosity gradient corresponding to the different sets of scale coefficients.
57
58
59 15
1
2
3 3.3 Controllability of manufacture precision by SLM
4
5
6
7 Figure 14 illustrates the porous structures of SLMed specimens and reconstructions by XCT.
8
9 All struts, both inside and outside were fabricated precisely. Table 4 lists the contrast between
10
11 the design porosity and the as-built porosity. The porosity deviations between the models and the
12
13
14
specimens were less than 4%. This met the requirements of high-precision manufacturing of
15
16 porous scaffolds.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49 Figure 14. Porous structures with different porosities and irregularities. (a)Porosity series,ε=0.41.
50
51 (b)Irregularity series, Φ=85.5%±2%.
52
53
54
Table 4. Porosity of CAD models and specimens of porous structure
55
56
57
58
59 16
1
2
3
Porosity/%
4 Series Deviation/%
5 Model Specimen
6
7 1 64.77 63.18 2.45
8
9 2 71.83 69.68 2.99
10
11 Porosity 3 78.46 75.57 3.69
12
13 Series 4 85.66 82.43 3.77
14
15 5 90.24 87.50 3.03
16
17 6 93.91 91.86 2.18
18
19
1 86.26 84.89 1.59
20 2 85.75 83.92 2.14
21
22 Irregularity 3 85.50 83.66 2.16
23
24 Series 4 85.44 84.05 1.63
25
26 5 85.50 83.83 1.96
27
28 6 85.69 83.67 2.36
29
30
31
32
33 3.4 Controllability of mechanical properties
34
35
36
37
Figure 15 shows the compressive curves of these two series. Figure 16 shows the the
38
39 relationship between porosity Φ, apparent elastic modulus E, and compressive strength S. Figure
40
41 17 shows the relation between irregularity, elastic modulus, and compressive strength.
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 17
1
2
3 Figure 15. The compressive curves. (a) Porosity series, ε=0.50. (b) Irregularity series, Φ=85.5%
4
5
6 ±2%.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25 Figure 16. Relation between porosity and (a) elastic modulus, and (b) compressive strength.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Figure 17. Relation between irregularity and (a) elastic modulus, and (b) compressive strength.
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 18
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25 Figure 18. The gradient pore structure and compressive test curve.
26
27
28 There was a power function relationship between the compressive strength and porosity, as
29
30
well as a strong linear relationship between apparent elastic modulus and porosity. With
31
32
33 application of the Ashby-Gibson model, the elastic modulus and compressive strength of the
34
35 porous structure were fitted as:
36
37
  1.54
 E  11.83(1  100 )
38
39
 (8)
40
41  S  1400.62(1   ) 2.38
42
 100
43
44
45 Due to the variation of irregularities, the elastic modulus showed a sharp downward trend.
46
47 Meanwhile, the compressive strength exhibited a wide range of fluctuation with the increase of
48
49
50 irregularity, which was similar with previous work. 10 As shown in Figure 13 (a), the stiffness of
51
52 porous structure is highly dependant on its porosity and irregularity. On one hand, the decrease
53
54 of porosity, caused by the decrease of strut diameters and controlled by K, resulted in the decay
55
56
57
58
59 19
1
2
3 of struts stiffness, which made the porous structure deformed more easily; On another hand, the
4
5
6 increase of irregularity make more struts change from vertical or horizontal to tilted position,
7
8 which also reduce the structure stiffness. The special mechanical behaviors of irregular porous
9
10 structures, reflected in Figure 17(b), were attributed to the increase of irregularity and the SLM
11
12
13
process. Previous study 10believed that small levels of irregularity resulted in a decrease in
14
15 strength that is caused by the initial destabilization of the unit cell on which the structure was
16
17 based. When the irregularity exceeded a certain value, a new stress balance was built and
18
19
compressive strength tended to be stable. In addition, over the SLM processing, the pores and
20
21
22 cracks occurred in specimens inevitably24-25. These cracks , related to the orientation of the struts ,
23
24 had a significant influence on the compressive strength26-27. This leaded to the fluctuation in
25
26 compressive strength test data as well.
27
28
29 When the irregularity of porous scaffolds is given, the mechanical properties are in accordance
30
31 with the Gibson-Ashby model. However, when the irregularity is accounted for, the Gibson-
32
33 Ashby model is no longer applicable. The Gibson-Ashby model is based on the regular
34
35
36
assumption, ignoring the influence of irregularity in principle. It cannot reflect the change trend
37
38 of the irregular effect on the mechanical properties. The results from the current study
39
40 demonstrate that when the influence of irregularity is considered, the mechanical properties of
41
42
porous structures should be at least three variable function parameters of the solid material
43
44
45 (porosity, irregularity, and mechanical). As a result, further studies should focus on the
46
47 mechanism of this phenomenon and explore more precise theoretical models.
48
49 It is worth mentioning that, the structures that we prepared and tested are less stiff than bone.
50
51
52 According to previous work2, the range of apparent elastic moduli of bone is 1-10 GPa. The
53
54 values depend on the sample and method strongly, and different researchers have different
55
56
57
58
59 20
1
2
3 results28. For example, patients with osteoporosis tend to have loose bones, low trabecular elastic
4
5
6 modulus. In this case, low apparent modulus of elasticity is significant and necessary. As shown
7
8 in Figure 16, adjusting K and R can change the stiffness of the structure. The stiffness of porous
9
10 structures increases with decreasing K. According to Figure 17, the stiffness of the structure is
11
12
13
higher when the irregularity is lower (ε=0.11) or higher (ε=0.50). In addition, by optimizing the
14
15 test methods and process parameters, the apparent elastic moduli of the porous structures can be
16
17 further enhanced.
18
19
Figure 18 shows the gradient pore structure and the compression test curve. These structures
20
21
22 were designed according to the parameters in Table 3 and Figure 4. The mechanical behavior of
23
24 the gradient pore structures are significantly different from uniform porous structures. In the low
25
26 strain stage, the compression curve shows a linear trend similar to a uniform random porous
27
28
29 structure. In platform stage, the stress platform is no longer horizontal distribution, but showing
30
31 upward trend. And as the gradient increases, the angle of the platform curve increases. This is
32
33 different from a uniform porous structure. This is because the porosity of the porous structure
34
35
36
continuously decreases in the height direction, and the struts of the pore gradually deformed and
37
38 failed in the direction of increasing porosity. The larger the porosity gradient, the greater the load
39
40 required for the destruction, which is in agreement with the trend disclosed in Figure 16 in the
41
42
manuscript. Gradient porous structure not only has potential application value in mechanics, but
43
44
45 also has broad application prospect in tissue engineering scaffold. A realistic application is that it
46
47 provides an integrated scaffold for studying the osteosynthesis capabilities of different pore sizes.
48
49
50
4. Conclusion
51
52
53
54
55
56
57
58
59 21
1
2
3 This study proposed a probability sphere method in order to construct irregular structures
4
5
6 based on the Voronoi-Tessellation method. The porous specimens were fabricated by SLM.
7
8 Their compressive behaviors were investigated. Our findings are summarized as follows:
9
10 (1) This method allowed us to obtain not only an accurate design of the specific porosity, as
11
12
13
well as the gradient distribution of porosity. Porous scaffolds with special porosities ranging
14
15 from 60% to 95 % and pore size ranging from 200 µm to 1200 µm were designed precisely and
16
17 conveniently. A porosity gradient ranging from 0.03 to 0.54 was obtained.
18
19
(2) The proposed irregular coefficient could control the irregularity of the porous structures
20
21
22 and achieve good accommodation and balance of ‘irregularity’ and ‘controllability’.
23
24 (3) The compressive properties were affected by both porosity and irregularity. The
25
26 compressive strength decayed at low levels of irregularity. The compressive strength was
27
28
29 enhanced at high levels, which enabled us to obtain porous structure with high strength but low
30
31 stiffness.
32
33 (4) The compressive properties of porous structures could be adjusted to the level of cortical
34
35
36
bone. The apparent elastic moduli of porous specimens had high irregularity (0.50) that varied
37
38 from 0.14 GPa to 2.37 GPa. The compressive strengths varied from 1.94 MPa to 116.61 MPa.
39
40 (5) The mechanical behavior of the gradient pore structures are significantly different from
41
42
uniform porous structures in the platform stage. The compressive stress platform of gradient pore
43
44
45 structure tends to rise. The larger the porosity gradient, the larger the angle of ascent.
46
47 *Corresponding Author
48
49
50 Phone:+86 025 84892520
51
52
53
54 Author Contributions
55
56
57
58
59 22
1
2
3 The manuscript was written through contributions of all authors. All authors have given approval
4
5
6 to the final version of the manuscript.
7
8
9 ‡These authors contributed equally.
10
11
12 Funding Sources
13
14
15 The authors would like to thank the Key Research and Development Plan of Jiangsu Province
16
17 (No., BE2016010-3), Science and Technology Support Program (The Industrial Part), Jiangsu
18
19 Provincial Department of Science and Technology of China (No., BE2014009-1), the Science
20
21
22 and Technology Supporting Plan (The Industrial Part) of Zhangjiagang (No., ZKG1615), the
23
24 Science and Technology Supporting Plan (Social Development Part) of Zhangjiagang (No.,
25
26 ZKS1606), the Nation Nature Science Fund of China (No., 51475238).
27
28
29
30 REFERENCES
31
32
33 1.Geetha, M.; Singh, A. K.; Asokamani R.; Gogia, A. K. Ti based biomaterials, the ultimate
34
35 choice for orthopaedic implants- A review. Prog. Mater. Sci. 2009, 54, 397-425.
36
37 DOI:10.1016/j.pmatsci.2008.06.004.
38
39
40
2. Choi, K.; Kuhn, J. L.; Ciarelli, M. J.; Goldstein, S. A. The elastic moduli of human
41
42
43 subchondral, trabecular, and cortical bone tissue and the size-dependency of cortical bone
44
45 modulus. J Biomech. 1990, 23, 1103-1113, DOI:10.1016/0021-9290(90)90003-L.
46
47
48 3. Habibovic, P.; Yuan, H.; van der Valk, C. M.; Meijer, G.; van Blitterswijk, C. A.; de Groot,
49
50
51
K. 3D microenvironment as essential element for osteoinduction by biomaterials. Biomaterials
52
53 2005, 26, 3565-3575. DOI:10.1016/j.biomaterials.2004.09.056.
54
55
56
57
58
59 23
1
2
3 4. Otsuki, B.; Takemoto, M.; Fujibayashi, S.; Neo, M.; Kokubo, T.; Nakamura, T. Pore throat
4
5
6 size and connectivity determine bone and tissue ingrowth into porous implants: Three-
7
8 dimensional micro-CT based structural analyses of porous bioactive titanium implants.
9
10 Biomaterials. 2006, 27, 5892-5900, DOI:10.1016/j.biomaterials.2006.08.013.
11
12
13 5. Mastrogiacomo, M.; Scaglione, S.; Martinetti, R.; Dolcini, L.; Beltrame, F.; Cancedda, R.;
14
15
16 Quarto, R. Role of scaffold internal structure on in vivo bone formation in macroporous calcium
17
18 phosphate bioceramics. Biomaterials. 2006, 27, 3230-3237.
19
20 DOI:10.1016/j.biomaterials.2006.01.031.
21
22
23
6. Wang, Z.; Wang, C.; Li, C.; Qin, Y.; Zhong, L.; Chen, B.; Li, Z.; Liu, H.; Chang, F.; Wang,
24
25
26 J. Analysis of factors influencing bone ingrowth into three-dimensional printed porous metal
27
28 scaffolds: A review. J. Alloys Compd. 2017, 717, 271-285, DOI:10.1016/j.jallcom.2017.05.079.
29
30
31 7. Liu, Y. J.; Li, X. P.; Zhang, L. C.; Sercombe, T. B. Processing and properties of
32
33
topologically optimised biomedical Ti-24Nb-4Zr-8Sn scaffolds manufactured by selective laser
34
35
36 melting. Mater. Sci. Eng., A. 2015, 642, 268-278. DOI:10.1016/j.msea.2015.06.088.
37
38
39 8. Gorgin Karaji, Z.; Speirs, M.; Dadbakhsh, S.; Kruth, J.; Weinans, H. I. E. D.; Zadpoor, A. A.;
40
41 Amin Yavari, S. I. E. D. Additively Manufactured and Surface Biofunctionalized Porous Nitinol.
42
43
44
ACS Appl. Mater. Interfaces. 2017, 9, 1293. DOI:10.1021/acsami.6b14026.
45
46
47 9. Wang, X.; Xu, S.; Zhou, S.; Xu, W.; Leary, M.; Choong, P.; Qian, M.; Brandt, M.; Xie, Y.
48
49 M. Topological design and additive manufacturing of porous metals for bone scaffolds and
50
51 orthopaedic implants: A review. Biomaterials. 2016, 83, 127-141.
52
53
54
DOI:10.1016/j.biomaterials.2016.01.012
55
56
57
58
59 24
1
2
3 10. Mullen, L.; Stamp, R. C.; Brooks, W. K.; Jones, E.; Sutcliffe, C. J., Selective Laser Melting:
4
5
6 a regular unit cell approach for the manufacture of porous, titanium, bone in-growth constructs,
7
8 suitable for orthopedic applications. J. Biomed. Mater. Res., Part B. 2009, 89, 325-334.
9
10 DOI:10.1002/jbm.b.31219
11
12
13 11. Yang, N.; Gao, L.; Zhou, K., Simple method to generate and fabricate stochastic porous
14
15
16 scaffolds. Mater. Sci. Eng., C. 2015, 56, 444-450, DOI:10.1016/j.msec.2015.06.039.
17
18
19 12. Gómez, S.; M.D. Vlad; López, J.; Fernández, E. Design and Properties of 3D Scaffolds for
20
21 Bone Tissue Engineering. Acta Biomater. 2016, 42:341-350. DOI:10.1016/j.actbio.2016.06.032.
22
23
24 13. Sogutlu, S.; Koc, B. Stochastic Modeling of Tissue Engineering Scaffolds with Varying
25
26
27
Porosity Levels. Computer-Aided Design and Applications. 2007, 4, 661-670.
28
29 DOI:10.1080/16864360.2007.10738500.
30
31
32 14. Cai, S.; Xi, J. A control approach for pore size distribution in the bone scaffold based on
33
34 the hexahedral mesh refinement. Computer-Aided Design. 2008, 40, 1040-1050.
35
36
37 DOI:10.1016/j.cad.2008.09.004.
38
39
40 15. Lal, P.; Sun, W. Computer modeling approach for microsphere-packed bone scaffold.
41
42 Computer-Aided Design. 2004, 36, 487-497. DOI:10.1016/S0010-4485(03)00134-9.
43
44
45 16. Aurenhammer F. Voronoi diagrams-a survey of a fundamental geometric data structure.
46
47
Acm Computing Surveys, 1991, 23,345-405. DOI:10.1145/116873.116880
48
49
50
51 17. Devillers, O.; Golin, M.; Kedem, K.; Schirra, S., Queries on Voronoi diagrams of moving
52
53 points. Computational Geometry. 1996, 6, 315-327. DOI:10.1016/0925-7721(95)00053-4.
54
55
56
57
58
59 25
1
2
3 18. Chow, H. N.; Tan, S. T.; Sze, W. S. Layered Modeling of Porous Structures with Voronoi
4
5
6 Diagrams. Computer-Aided Design and Applications. 2007, 4, 321-330.
7
8 DOI:10.1080/16864360.2007.10738552.
9
10
11 19. Wejrzanowski, T.; Skibinski, J.; Szumbarski, J.; Kurzydlowski, K. J. Structure of foams
12
13 modeled by Laguerre-Voronoi tessellations. Comput. Mater. Sci. 2013, 67, 216-221.
14
15
16 DOI:10.1016/j.commatsci.2012.08.046.
17
18
19 20. Randrianalisoa, J.;Baillis, D.; Martin, C. L; Dendievel, R. Microstructure effects on
20
21 thermal conductivity of open-cell foams generated from the Laguerre-Voronoï tessellation
22
23
method. Int. J. Therm. Sci. 2015, 98:277-286. DOI:10.1016/j.ijthermalsci.2015.07.016.
24
25
26
27
21. Xiao, F.; Yin, X. Geometry models of porous media based on Voronoi tessellations and
28
29 their porosity – permeability relations. Computers & Mathematics with Applications. 2016, 72,
30
31 328-348, DOI:10.1016/j.camwa.2015.09.009.
32
33
34 22. Gibson, L. J. A.A.; Ashby, M. F. Cellular Solids: Structure And Properties. Cambridge
35
36
37 University Press: Cambridge, 2014, 33; pp 487-488, DOI:10.1017/CBO9781139878326.
38
39
40 23. Stamboulis, A. G.; Boccaccini, A. R.; Hench, L. L. Novel Biodegradable
41
42 Polymer/Bioactive Glass Composites for Tissue Engineering Applications. Adv. Eng. Mater.
43
44 2002, 4:105-109, DOI: 10.1002/1527-2648(200203).
45
46
47
24. Craeghs, T.; Clijsters, S.; Yasa, E.; Bechmann, F.; Berumen, S.; Kruth, J., Determination of
48
49
50 geometrical factors in Layerwise Laser Melting using optical process monitoring. Optics and
51
52 Lasers in Engineering. 2011, 49, 1440-1446. DOI:10.1016/j.optlaseng.2011.06.016.
53
54
55
56
57
58
59 26
1
2
3 25. Wauthle, R.; Vrancken, B.; Beynaerts, B.; Jorissen, K.; Schrooten, J.; Kruth, J.; Van
4
5
6 Humbeeck, J. Effects of build orientation and heat treatment on the microstructure and
7
8 mechanical properties of selective laser melted Ti6Al4V lattice structures. Addit. Manuf. 2015, 5,
9
10 77-84. DOI:10.1016/j.addma.2014.12.008
11
12
13 26. Popovich, A.; Sufiiarov, V.; Borisov, E.; Polozov, I. Microstructure and mechanical
14
15
16 properties of Ti-6Al-4V manufactured by SLM. Key Eng. Mater. 2015, 651-653, 677-682.
17
18 DOI:10.4028/www.scientific.net/KEM.651-653.677
19
20
21 27. Yang, Y.; Wu, P.; Lin, X.; Liu, Y.; Bian, H.; Zhou, Y.; Gao, C.; Shuai, C. System
22
23
development, formability quality and microstructure evolution of selective laser-melted
24
25
26 magnesium. Virtual & Physical Prototyping. 2016, 11, 173-181,
27
28 DOI:10.1080/17452759.2016.1210522
29
30
31 28. Boruah, S.; Subit, D. L.; Paskoff, G. R.; Shender, B. S.; Crandall, J. R.; Salzar, R. S.
32
33
Influence of bone microstructure on the mechanical properties of skull cortical bone – A
34
35
36 combined experimental and computational approach. J. Mech. Behav. Biomed. Mater. 2017, 65,
37
38 688-704. DOI:10.1016/j.jmbbm.2016.09.041.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 27
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15 For Table of Contents Use Only
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 28

15 Medical Voroni Niti

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

15 Medical Voroni Niti

Uploaded by

Copyright:

Available Formats

Subscriber access provided by READING UNIV

ACS Biomaterials Science & Engineering is published by the American Chemical

You might also like