Mechanical properties of porous titanium alloy scaffold fabricated using additive manufacturing technology

Abstract

Additive manufacturing technology has been widely used for the fabrication of porous titanium (Ti) orthopedic implants. Though high porosities and low stiffness of the scaffolds are benefit for bone ingrowth, the compressive strength and fatigue performance of the scaffolds are still necessary for load-bearing applications. In this study, a novel porous scaffold with framework structure was designed and fabricated using selective laser melting process. The mechanical properties of the porous Ti scaffolds were evaluated. The results showed that the compressive strength of the scaffolds possessing framework structure was enhanced significantly with the porosity decreased slightly. In addition to static mechanical properties, the fatigue properties of the porous Ti scaffolds were also improved via the framework structure.

Keywords

Additive manufacturing porous titanium compressive strength fatigue test framework structure

1. Introduction

In recent years, orthopedic diseases are becoming more and more common, especially with the aging of population [1]. Orthopedic implant/scaffold is the main means of bone repair for bone defects. Titanium and its alloy are the most widely used kind of material to repair load-bearing defects [2]. However a traditional bulk of titanium or its alloy can cause stress shielding because of the stiffness mismatch, leading to aseptic loosening of scaffold and osseointegration failure [3–5]. Besides, the surface of the solid scaffold is smooth and has no pores that allow human skeletal cells to grow into. As a result, human bone cells can only grow on the surface with less strong bone cell adhesion [6].

The arrival of additive manufacturing (AM) technologies, selective laser sintering (SLS) for example, has showed the possibility to solve this problem [7–9]. With AM technology, researchers can fabricate porous titanium scaffolds with desired architecture, controllable porosities, and proper stiffness [10]. Porous titanium have advantages in promoting bone ingrowth with better biomechanical compatibility [11].

Researchers usually hope the porosity of the porous scaffold is the higher the better for better osseointegration and bone regeneration if ensuring the strength requirements. Requirement on the pore size of the scaffold is 200–800 μm, since pores within this size range are more suitable for cell growth. Porosity of human cancellous bone is 30–95%; for cortical bone it is 5.36–14.2% [12]. The elastic modulus of scaffolds is required to be the same as that of human’s bone. However, the porosity of scaffold has negative effect on the compressive strength. How to obtain good compressive strength with high porosity is still a problem [13]. To solve it, we enlarged the porosity of scaffolds and maintained the stuffiness as human bone by adding framework structure in the design in a recent novel design.

In this study, selective laser melting (SLM) technique was used to fabricate porous Ti scaffolds with framework structure. The mechanical characteristics of samples were evaluated and compared through microscope observation test, compressive strength test and fatigue properties test. The static mechanical properties and fatigue properties of the fabricated Ti scaffolds with framework structure were examined so that the feasibility of this new design and the manufacturing process will be proved in this paper.

2. Materials and methods

2.1. Design and manufacturing

Commercial CAD software Unigraphics NX 10.0 was used to design the porous model (Fig. 1). A variety of diamond lattice units piled up, forming the spatial arrangement of a basic array. Using parametric structural design method, the porosity of the array could be changed via adjusting the unit’s size, including width (W), length (L) and height (Z) (Table 1). Then the array was performed a Boolean summation with the target model (here is a cylinder with 10 mm diameter and 13 mm height) to make it porous. The three dimensions of the bar size, W, L, and Z, determined the design porosity of the model. However, there is only a qualitative relationship between rod size and designed porosity (P_d) due to the limited and uncertain cylindrical boundaries [14]: $\begin{eqnarray}\displaystyle P_{d}=\frac{V_{1}-V_{2}}{V_{1}}\times 100\%. & & \displaystyle\end{eqnarray}$ (1) In which, P_d is the design porosity, V₁ is the model’s volume (cylinder), V₂ is the volumn of porous model calculated by software.

Table 1

The size and static mechanical properties of the scaffold series

Series No.	Parameter/mm	Diameter of	Porosity	Porosity (dry	Compression	Elastic
		the framework	(designed)/%	weighing)/%	Strength /Mpa	Modulus/Gpa
		structure/mm
1	0.4-0.8-0.5	0	66.4	69.5	114.8	3.26
2	0.2-0.5-0.5	0	78.2	81.8	41.4	1.89
3	0.2-0.5-0.5	1	71.9	74.5	103.3	3.43
4	0.2-0.5-0.5	1.5	67.1	68.9	188.5	5.30

Then the CAD data was converted into the input file for SLM machine (M2, Concept Laser, Germany) to be fabricated into porous Ti scaffolds. The raw material used in the additive manufacturing process was Ti6Al4V (Ti) powder, whose diameter was about 15–45 μm. The framework structure was designed and Boolean summed to the basic porous model to improve the mechanical properties of the scaffolds. The diameter of the framework’s bar could be adjusted to modify the mechanical performance. The additional framework structure consisted mainly of five vertical cylindrical bars and several horizontal bars on both upper and lower surfaces. The horizontal bars were connected to the vertical bars (Fig. 1).

After the samples were manufactured, the surface morphology and the micro-structure were observed by optical microscope (VHX-500F, KEYENCE, Japan). The diameter of the pores and the bars were measured by the measurement toolbox in the microscope system. Dry weighing porosity calculation was conducted by dividing the actual weight of one sample by the theoretical weight of the cylinder Ti6Al4V model with the density of 4.42 g/cm³.

2.2. Mechanical tests

The static compression tests of the porous Ti scaffolds were performed using a material testing system (MTS 810, USA) with a 10 kN load cell, and a 0.5 mm/min cross-head speed was set at room temperature. The stress calculated by dividing the compression fracture force P_b by the initial cross-sectional area of the sample was the compression strength σ_b, and the slope of the compressive stress-strain curve in the linear region was the elastic modulus E.

Fig. 1.

Series of porous scaffolds (a) 0.4-0.8-0.5 without framework (b) 0.2-0.5-0.5 without framework (c) 0.2-0.5-0.5 with 1 mm framework (d) 0.2-0.5-0.5 with 1.5 mm framework.

Fatigue tests were performed on electronic fatigue testing system (E10000, Instron, USA) (Fig. 2). For every scaffold series, there were three fatigue testing samples. About 0.25σ_b or 0.2σ_b was chosen as the first load stress (load force divided by the initial cross-sectional area of the sample), and about 0.1σ_b or 0.05σ_b was chosen as the fourth load stress. Here σ_b was the ultimate compression strength of the sample got in the static compression test. Another two test points were inserted into the former two points then. The value of four test loads formed an arithmetic sequence. To simplify the configuration of fatigue test system, the value of load compression force were set as integers (Table 2).

Fig. 2.

Fatigue test system and the test scaffold series.

Table 2

The force values loaded on samples in fatigue tests of scaffold series

		Test point 1		Test point 2		Test point 3		Test point 4
Series	Number of	Load	σ∕σ_b	Load	σ∕σ_b	Load	σ∕σ_b	Load	σ∕σ_b
No.	test samples	Force/N		Force/N		Force/N		Force/N
1	3	2200	0.244	1800	0.200	1400	0.155	1000	0.111
2	3	700	0.215	600	0.185	500	0.150	400	0.123
3	3	1600	0.197	1200	0.148	800	0.099	400	0.049
4	3	3000	0.203	2000	0.135	1500	0.101	1000	0.068

Fatigue test parameters were as follows: a sine wave load waveform at 5 Hz, 0.1 for the load ratio (namely 0.9 test load at troughs and 1.1 test load at crests), and 3 million times for the test cycle at normal temperature and pressure. The fixture was two polyformaldehyde (POM) blocks. The collected data was analyzed intending to find the relationship between controlling variables and mechanical properties of the four series. The fatigue test data was imported into Origin Pro 2017 and analyzed [15].

3. Results

3.1. Characterization of scaffolds

Figure 1 shows the fabricated porous Ti scaffolds series samples. Table 1 is the results of the static compression test, which also includes the design parameters of the four series. For series 1, the unit’s width was 0.4 mm, the unit’s length is 0.8 mm and the unit’s height is 0.5 mm. For series 2& series 3& series 4, the unit’s three parameter were all 0.2 mm, 0.5 mm and 0.5 mm, while the diameter of the framework structure of series 3 was 1 mm and that of series 4 was 1.5 mm. The real porosity of scaffold samples were all bigger than the designed porosity through the try weighing.

The compressive strength of series 2, series 3 and series 4 increased as the framework structure became thicker and the porosity decreased. Usually the porosity has a direct influence on the mechanical properties: scaffold with larger porosity has lower strength. Howerer, the porosity of series 4 was nearly the same as that of series 1, but its compression strength was much larger (Table 1). In other words, the 1.5 mm diameter framework structure was able to raise the strength by more than a half without degrading the porosity. The static compressive strength was significantly improved by framework structure with the porosity decreased slightly. The elastic modulus was 3.43–5.30 Gpa, which was still close to that of human’s cortical bone (The elastic modulus of the human cancellous bone is about 0.3–1 GPa, and the elastic modulus of the cortical bone is about 10–20 GPa).

Table 2 listed the force loaded on the samples during the fatigue test. For series 1, the compression force changed from 2200N to 1000N, including four test points making up an arithmetic sequence. For every test point, three samples were tested. Several similar sets of points were applied on the samples of series 2, series 3 and series 4. There was one more test point at 2500N for series 4 which was not listed here or included in the data analysis process.

The surface morphology of series 1 and series 2 was illustrated in Fig. 3. Due to the drawbacks of the manufacturing process, the slabs of the SLM-made samples were uneven. Many metal particles were adsorbed on the surface of the rods. The width and the pore size of the rods of the porous titanium sample were measured under an optical microscope and it was found that the measured values were smaller than the designed values. This also explained why the actual porosity was larger than the designed.

Fig. 3.

Surface morphology of series 1 (left) & series 2 (right).

3.2. Fatigue behavior

Figure 4 illustrates the fatigue test results of all four series. The designed parameters of series 2, series 3 and series 4 were the same. The only difference was the diameter of framework structure. As the diameter became larger, the S-N curve moved up, meaning that adding framework structure made cycles to failure increase under the same test load force. It should be noted that, as the framework structure became thicker, the porosity was also slightly reduced. The series 1 was tested as a reference. Compared with the sample of series 1, the fatigue performance of series 4 was significantly improved by framework structure so that the sample could also sustain more than 3 million test cycles under 1000N load. It sustained over 1 million cycles even under 1500N load. This novel design satisfied most load-bearing applications under the premise of greatly improved porosity.

Fig. 4.

S-N curves of four fatigue series with power law fitting curves.

Data of S-N curves in Fig. 4 was normalized by dividing the test stress by the compression stress of the series. Before being normalized, the stress values could all be fitting to power law with the coefficient of determination R ², larger than 0.9. Figure 5 illustrates the relationship between normalized stress of all four series and the cycles to failure. A power law curve was calculated to fit this relationship with the high coefficient of determination R ² = 0.92, that is: $\begin{eqnarray}\displaystyle {\sigma}/{\sigma}_{b}=3.503N^{-0.232}. & & \displaystyle\end{eqnarray}$ (2)

4. Discussion

The SLM manufacturing process will leave some metal powder attached to the pores of the sample, which will have a bad effect on the quality of the scaffold no matter for the surface morphology or the mechanical test. In addition to the residual powder in pores, there are some semi-molten particles, which were also caused by the manufacturing process itself. These metal particles were the semi-molten metal particles making the surface of the sample rough (Fig. 3). And there were a lot of small internal angles inside porous titanium scaffold. These internal angles were difficult to be accurately printed out by 3D printer. Therefore, it was observed that the measured parameter was deviated from the designed value.

Fig. 5.

Normalized S-N curve of all fatigue samples with one power law fitting curve.

The S-N curve (Fig. 4) can predict when the sample will be fractured during fatigue test for the four series under different loads. The normalized S-N curve can be used to predict the corresponding S-N curve for other series as the fatigue test is too time consuming and costly. It is interesting that the normalized stress has a power-law relation with the cycle number. If the cylindrical porous titanium is designed and manufactured using different parameters, normalized S-N curve in Fig. 5 and the function (2) can be used to obtain the corresponding series’ S-N curve.

After the verification of this study, the design of the framework structure can greatly improve the mechanical properties of the scaffold under the premise of increasing the porosity, whose elastic modulus still matches human bone, avoiding stress shielding effect.

Unlike the biomaterials that can be absorbed by human body [16], the designed scaffolds here are fabricated by titanium and are not resorbable. They are used as a substitute for hard tissues on which bone cells colonize. Pores inside them are channels for cell growth. That’s why the porosity of the scaffold is crucial for their osseointegration capability [17]. The framework structure can increase the porosity which motivates more new bone formation in bone graft area.

Compared with the common designed structure [18], the framework structure also improves the mechanical properties such as compression strength, elastic modulus and fatigue performance of porous titanium orthopedic implants. Fatigue performance has been greatly improved. It ensures the implant can withstand a longer period of fatigue loading without failure after being implanted into the human body. In some implants that require strong osseointegration and longer fatigue life, taking femoral stems, acetabular cups, and intervertebral cages as an example, the designed framwork structure can greatly increase the service life and biocompatibility of the implant.

Since the scaffold usually sustain the pressure from the certain direction, the framework structure is designed to reinforce the scaffold’s certain side. The mechanical performance test only includes compression test and low frequency (5 Hz) compression fatigue test without the tensile test, the shear force test, the torsion test, etc., because here we only intend to verify the effect of framework structure design. It is certain that since the direction of the support structure is in the axial direction of the cylinder, this design has a significant advantage in bearing axial tension. For different parts of the human bone, the biomechanical performance varies widely so that the direction of force and fatigue are very different. The designed structure in this study can also be used as a micro-unit, and then many cylindrical porous titanium combine to form a prosthesis. In this process, the arrangement of the cylinders can be varied and the cylinders need to be arranged according to the mechanical properties of the prosthesis.

In fact, this design method, which is validated by mechanical tests, illustrates the design optimization in macroscopic dimensions for better mechanical properties. Adding framework structure is just one design method that provides an inspiration that the porous titanium scaffold can be designed in a gradient structure. The implant is not composed of uniform micro units, but high-density units at stress concentrated positions and low-density units at positions bearing small force. This structure can be named as artificial bone as it imitates the human bone structure with cortical bone and cancellous bone.

5. Conclusion

Through static and dynamic compression test, the main mechanical properties of the scaffolds samples were evaluated. It is found that although the porosity has a direct influence on the mechanical properties, the framework structure is able to raise the static compressive strength significantly with the porosity decreased slightly. Through the dynamic compression test, the fatigue performance of series is proved to be strengthened by framework structure. This novel design satisfies most load-bearing applications under the premise of greatly improved porosity. The normalized S-N curve conforming to a power law can be used to predict the corresponding S-N curve for other series using the same manufacturing process. Under the guidance of the presented design method, it is possible to manufacture the porous titanium implant meeting the demands of biomechanical properties in a larger range.

However the design process introduced in this paper is rather empirical. In future, more theoretical design ideas can be utlized to optimize the framework structure in the design to further improve the mechanical performance of porous titanium orthopedic implants. For example, it is a good way to design gradient porous structure for artificial bone based on unit’s topology optimization methodology.

Footnotes

Acknowledgements

The autours gratefully acknowledge the funding support from the National Natural Science Foundation of China (Grant No. 51475293).

References

Demir

Gürsu

Yildirim

, Evaluation of anxiety levels in patients with chronic orthopedic diseases, Acta Orthopaedica Et Traumatologica Turcica 46(6) (2012), 420.

Saiz

and Tomsia

A.P.

, Direct ink writing of highly porous and strong glass scaffolds for load-bearing bone defects repair and regeneration, Acta Biomaterialia 7(10) (2011), 3547–3554.

Ben-Nissan

and Pezzotti

, Bioceramics: Processing routes and mechanical evaluation, Journal of the Ceramic Society of Japan 110(1283) (2002), 601–608.

Noyama

Miura

Ishimoto

, Bone loss and degradation of bone quality in the human femur after total hip arthroplasty under stress-shielding by titanium-based implant, Journal of the Japan Institute of Metals 76(7) (2012), 468–473.

Yang

Chen

H.J.

Zhu

X.D.

, Enhanced repair of a critical-sized segmental bone defect in rabbit femur by surface microstructured porous titanium, Journal of Materials Science Materials in Medicine 25(7) (2014), 1747–1756.

Turner

M.L.

Anthony

R.S.

Parker

A.C.

, Differential expression of cell adhesion molecules by human hematopoietic progenitors from bone marrow and mobilized adult peripheral blood, Stem Cells 13(3) (2010), 311–316.

Hoffmann

Bormann

Rossi

, Rapid prototyped porous nickel-titanium scaffolds as bone substitutes, Journal of Tissue Engineering 5(5) (2014), 1–14.

Wysocki

Święszkowski

and Kurzydłowski

K.J.

, Porous titanium scaffold fabrication for bone tissue engineering by SLM (selective laser melting) method, in: Ddmc 2014 Fraunhofer Direct Digital Manufacturing Conference, 2014.

Inga

Liga

Janis

, Development and modification of novel 3D porous titanium dioxide scaffolds for bone tissue regeneration, Frontiers in Bioengineering & Biotechnology 4 (2016).

10.

Deville

Meille

and Seuba

, A meta-analysis of the mechanical properties of ice-templated ceramics and metals, Science & Technology of Advanced Materials 16(4) (2015).

11.

Fan

Feng

Liu

, Fabrication of TiO2 nanotubes on porous titanium scaffold and biocompatibility evaluation in vitro and in vivo, Journal of Biomedical Materials Research Part A 100A(12) (2015), 3422–3427.

12.

G.I.

J.Y.

and Lee

J.H.

, Chondrogenesis of adipose stem cells in a porous polymer scaffold: Influence of the pore size, Cell Transplantation 21(11) (2012), 2397–405.

13.

X.Y.

Feng

Y.F.

, A novel composite scaffold consisted of porous titanium and chitosan sponge for load-bearing applications: Fabrication, characterization and cellular activity, Composites Science & Technology 117 (2015), 78–84.

14.

Greitemeier

Palm

Syassen

, Fatigue performance of additive manufactured TiAl6V4 using electron and laser beam melting, International Journal of Fatigue 94 (2016), 211–217.

15.

Van

H.B.

Apers

Lietaert

, Improving the fatigue performance of porous metallic biomaterials produced by Selective Laser Melting, Acta Biomaterialia 47 (2017), 193.

16.

Lekszycki

and Dell’Isola

, A mixture model with evolving mass densities for describing synthesis and resorption phenomena in bones reconstructed with bio-resorbable materials, ZAMM - Journal of Applied Mathematics and Mechanics 92(6) (2012), 426–444.

17.

Giorgio

Andreaus

Dell’Isola

, Viscous second gradient porous materials for bones reconstructed with bio-resorbable grafts, Extreme Mechanics Letters 13 (2017), 141–147.

18.

Spagnuolo

Barcz

Pfaff

, Qualitative pivot damage analysis in aluminum printed pantographic sheets: Numerics and experiments, Mechanics Research Communications 83 (2017).