Bioactive Materials 4 (2019) 56–70

Contents lists available at ScienceDirect

Bioactive Materials
journal homepage: http://www.keaipublishing.com/biomat

Additive manufacturing technology for porous metal implant applications T

and triple minimal surface structures: A review
Li Yuan, Songlin Ding∗∗, Cuie Wen∗
School of Engineering, RMIT University, Bundoora, Victoria, 3083, Australia

A R T I C LE I N FO A B S T R A C T

Keywords: Recently, the fabrication methods of orthopedic implants and devices have been greatly developed. Additive
Additive manufacturing manufacturing technology allows the production of complex structures with bio-mimicry features, and has the
Powder bed fusion potential to overcome the limitations of conventional fabrication methods. This review explores open-cellular
Porosity structural design for porous metal implant applications, in relation to the mechanical properties, biocompat-
TPMS structures
ibility, and biodegradability. Several types of additive manufacturing techniques including selective laser sin-
tering, selective laser melting, and electron beam melting, are discussed for different applications. Additive
manufacturing through powder bed fusion shows great potential for the fabrication of high-quality porous metal
implants. However, the powder bed fusion technique still faces two major challenges: it is high cost and time-
consuming. In addition, triply periodic minimal surface (TPMS) structures are also analyzed in this paper, tar-
geting the design of metal implants with an enhanced biomorphic environment.

1. Introduction improved in the last twenty years.

Materials for medical applications need to meet several criteria, and
Bones fundamentally provide frames for skeleton structural support, designed implants should morphologically mimic bone structure and
protection for vital organs and specialized tissues, support for the me- support bone tissue formation (osteogenesis). Biocompatibility, me-
chanical actions of soft tissues, and stability, and also play the role of chanical properties, and biodegradability are the fundamental elements
mineral storage. Bones have excellent regenerative properties and self- that must be considered. The structure of bone is almost completely
healing abilities for the body to recover from physical injury. However, constituted by hydroxyapatite crystal (Ca10 (PO4 )6 (OH )2 ), a mineral
after a serious trauma or a systemic disease, bones may find it ex- form of calcium apatite, within an organic matrix of collagen [4]. Of
tremely difficult to recover their self-healing function. Furthermore, this collagen, 95% is type I, providing the structural integrity for con-
bone's regenerative ability becomes weaker for older people, bringing a nective tissues in bones, tendons, and ligaments. The remaining 5% of
significant health issue of bone defects. The needs for orthopedic im- the bone is a combination of proteoglycans and numerous non-col-
plants have dramatically increased in the last two decades. Patients are lagenous proteins. An effective implant will be accepted by the human
expecting treatments that allow them to maintain their daily activities body and function properly. An inferior orthopedic device can trigger
and quality of life. Taking Knee replacement as an example, the rates of serious issues in patients. In Australia, the Therapeutic Goods Admin-
total knee replacement (TKR) have been significantly increased from istration (TGA) provides regulations for medical devices, such as Aus-
1991 to 2006 [1]. In the US, patients spent over US$9 billion on the tralian regulatory guidelines for medical devices (ARGMD), which
total knee arthroplasty (TKA) in 2009 and the major demand of TKA provide guidance to assist the manufacturers and sponsors of medical
was in group aged 45–64 years [2]. The value of the biomaterials devices in meeting the regulatory requirements for legally supplying a
market was US$94.1 billion in 2012, and increased to US$134.3 billion medical device in Australia [5,6]. The choice of implant materials
in 2017 worldwide [3]. This rapid increase in the biomaterials market should meet the requirements of the specific implant to ensure non-
has to a certain extent brought benefits for the development of bone toxicity and biocompatibility, a porous structure with appropriate pore
tissue engineering (BTE). Advanced biomaterials, fabrication methods, size and porosity, suitable biomechanical properties including appro-
and the structural designs of medical devices have been greatly priate elastic modulus and high strength, and also biodegradability for

Peer review under responsibility of KeAi Communications Co., Ltd.

∗
Corresponding author.
∗∗
Corresponding author.
E-mail address: [email protected] (C. Wen).

https://doi.org/10.1016/j.bioactmat.2018.12.003
Received 7 November 2018; Received in revised form 16 December 2018; Accepted 16 December 2018
2452-199X/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

temporary implant materials. 1.3. Biodegradability for temporary implant materials

In the development of new metallic biomaterials, three important
aspects must be considered: Recently, the fabrication of metal implants with an open-cellular
structure using advanced additive manufacturing technology has
mostly been able to meet the requirements for low elastic modulus, new
1.1. Non-toxicity and biocompatibility bone tissue ingrowth, and vascularization capacity for implantation
purposes. However, there are still concerns about the long-term in-
A clinical-level orthopedic implant has to be highly biocompatible stability of implants. Specifically, metallic biomaterials such as Ti and
and contain no toxic substances, ensuring the cell activities are safe and its alloys do not degrade over time, which means they remain perma-
positive. nently in the human body as a foreign object. As a consequence, in-
Metallic materials for implantation purposes should be non-toxic. A fections due to bacteria, long-term endothelial dysfunction, permanent
chemical substance or a mixture of substances can be poisonous, re- physical irritation, and local chronic inflammatory reactions can occur
sulting in adverse effects on living cells and organisms. Toxicity de- in patient and affect their quality of life [22]. In some cases, medical
scribes the level of a toxin in a substance [7,8]. Metals and their implants made of non-biodegradable materials like Ti alloys and
compounds can be toxic or cytotoxic. Most heavy metals, such as stainless steels can arrest the natural growth of bones, so that secondary
mercury (Hg), lead (Pb), and cadmium (Cd), are a major threat to surgery is necessary to support further bone ingrowth. As such, devel-
human life, damaging organs and cells [9]. Some metals are essential oping biodegradable metals for implant applications could be a better
minerals for the human body, such as barium (Ba), chromium (Cr), iron solution to these problems. The degradable implants will finally dis-
(Fe), selenium (Se), beryllium (Be), cobalt (Co), lithium (Li), strontium appear and be displaced by newly formed tissues. Biodegradability, also
(Sr), boron (b), copper (Cu), molybdenum (Mo), tungsten (W), cerium known as bio-absorbability, is considered for temporary medical im-
(Ce), iodine (I), nickel (Ni), and zinc (Zn) [10]. However, excessive plant application for bone fixation and vascular stents, such as implants
amounts of these essential minerals are toxic [11]. Furthermore, metal using magnesium (Mg)-, iron (Fe)-, and Zn-based alloys. The degrada-
compounds also have potential health risks, as they are prone to che- tion rate is an important factor for biodegradable implants [23]. The
mically breaking down in aqueous environments (hydrolysis). After a degradation of Zn-based alloys has been reported as the most appro-
chemical reaction, such compounds may release toxic compounds or priate among the biodegradable metal materials, in the range of
form insoluble residuals [12]. 20–300 μm/year in vitro [23–25]. The degradation rates of Mg-based
Biocompatibility describes a complex characteristic of a system alloys and Fe-based alloys are generally higher than 300 μm/year and
between the material and the biological host, rather than a simple lower than 50 μm/year in vitro, respectively [26,27]. Bowen et al. [28]
property of a material. The interaction between a biocompatible ma- investigated the corrosion behavior of pure Zn wire as cardiac stents in
terial and living cells or tissues should be positive, guiding wound rats for a six-month period of time. In their findings, the cross-sectional
healing, reconstruction and tissue integration. Therefore, the inter- area reduction of the stents (i.e., the corrosion rate) averaged 20 μm/
pretation of the concept of biocompatibility has escalated from being year for the first three months and more than doubled by the sixth
equivalent to non-toxic to have positive effect on interacting with living month. During immersion tests, those biodegradable metallic materials
cell and actively expressing new tissue [13]. Metal implants should be fabricated by AM techniques also provide mechanical properties ade-
biocompatible, thus providing a supportive environment for cell and quate for implant applications. Li et al. [29] fabricated an Fe-based
tissue activities. For instance, scaffolds in tissue engineering act as a diamond lattice scaffold by selective laser melting (SLM). In a four-
template for osteogenesis and the materials of scaffolds should be week immerse test, they found that the mechanical properties of the
compatible with the primary bone cells (osteoblasts, osteocytes, and scaffold were still sufficient for implants applications, where the elastic
osteoclasts), facilitating new bone formation, remodeling, and healing modulus and the yield strength of the samples were slightly decreased
[14–16]. by 7% and 5%, respectively.

2. Porous structure and porosity of implant materials

1.2. Biomechanical properties
The definition of porosity is the percentage of void space in a solid
Metallic biomaterials such as titanium (Ti) and some of its alloys are structure [30]. Porosity (P) can be calculated by the gravimetric
widely used in load-bearing implant applications due to their highly method, given by:
desirable mechanical properties, including high strength, relatively
P = (1 − ρstructure /ρmaterial)⋅100% (1)
suitable elastic modulus, fracture toughness and fatigue strength [17].
High-strength implants support physical activities and protect patients where ρmaterial is the density of the bulk alloy and ρstructural is the density
from fracture risks [18]. Implants with an appropriate elastic modulus of the porous structure, which is calculated as the mass of the structure
can prevent stress-shielding. A significant mismatch between the elastic (m) divided by its volume (V) [31]. In this paper, the analyses are based
moduli of an orthopedic implant and its surrounding bone can trigger on the desired porous structure, irrespective of porosity defects.
the so-called stress shielding, which occurs when the physical stresses
are taken up by the implant rather than by the bone. Stress shielding is 2.1. Porous structure with appropriate pore size and porosity
a cause of bone atrophy and may lead to implants loosening, and
eventually premature failure of the implant [19]. The elastic modulus Bone is an open-cell, porous composite material laid down by os-
of cancellous bone is in the range of 22.4–132.32 MPa, whereas that of teoblast cells and can be classified into two categories: compact bone
cortical bone is much higher, ranging from 7.7 to 21.8 GPa [20,21]. and cancellous bone. The former is the hard outer shell of bone with
Metal implants should exhibit an elastic modulus mimicking that of lower porosity [32], while the latter has bone cells interconnecting to
natural human bone. However, the elastic modulus of metallic implants form a highly porous structure that is lighter and more delicate than the
normally exceeds that of bone; for instance, the elastic modulus of outer shell of bone. The porosity of cancellous bone ranges from 30% to
commercially pure (CP)eTi and Tie6Ale4V is around 112 GPa and 95% and the pore size ranges from 200 to 1000 μm [32–35]. Recently,
115 GPa, respectively, much higher than that of cortical bone. Thus, there has been increasing interest in the development of porous bone
reducing the elastic modulus to an appropriate value is important for implant materials that exhibit an architecture mimicking that of human
implant design. bone, because the porous structure can provide space for new bone
tissue ingrowth and body fluid circulation. Moreover, the porous

57
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

scaffolds can provide the necessary support for cells to proliferate and be considered in orthopedic implant design. To a great extent, an in-
maintain their differentiated functions, and their structure defines the crease in the porosity of a scaffold results in better biocompatibility.
ultimate shape of the new bone created during growth processes. However, this may also reduce the mechanical properties of the porous
structure. Porosity is generally shown to be inversely proportional to
2.2. Effect of porosity on biocompatibility the mechanical properties. In some cases, an increase in pore size can
also reduce the Young's modulus and yield strength, due to wall
Biomedical implants and devices with high levels of porosity have thickness thinning [45]. Balancing the biocompatibility with the me-
attracted interest because of their notable biocompatibility, which is chanical properties is a prerequisite in the development of a superior
beneficial for bone regeneration and formation [36,37]. Porosity is an biomaterial device. An adult human has 206 separate bones in total,
essential element of bone ingrowth in metal implants. In an early study, supporting their daily physical activities. In order to facilitate multiple
Kuboki et al. [38] investigated the osteogenesis induced by bone mor- human activities, bones need to withstand different types of mechanical
phogenetic protein using hydroxyapatite with both solid and porous loads, which can be quantified as force, stress, and strain [46]. Me-
particles in rats. Their results indicated that osteogenesis only occurs in chanical properties can be described and measured in several different
porous structures [32,38]. A satisfactory porous structure needs to be ways, which include tensile strength, compressive strength, hardness,
open-cellular and interconnected, which is important for cell distribu- elasticity of stiffness, plasticity, ductility, and toughness. All of these
tion and migration, thus facilitating blood vessel formation [39]. Due to mechanical properties need to be evaluated and analyzed during im-
cell size, the minimum pore size is required to be 100 μm, which allows plant production.
the bone cells to migrate and be transported for osteogenesis [39,40]. As mentioned above, Ti and some of its alloys (e.g., Tie6Ale4V) are
Implants with pores of 200–350 μm size support new bone and capillary widely used as load-bearing implants. However, they have a relatively
formation, facilitating bone ingrowth [32]. In order to achieve better higher Young's modulus than human bone. A higher Young's modulus in
bone tissue ingrowth in scaffolding applications, the level of porosity implant materials than that of bone can cause bone atrophy due to a
was suggested to be 60% of the entire architecture as a minimum re- stress-shielding effect. How to minimize the Young's modulus is one of
quirement, with a pore size ranging from 200 to 1200 μm [20]. Torres- the major concerns in implant design. To address this issue, a me-
Sanchez et al. [41] studied the biological behaviors of porous Ti scaf- tallurgical approach can be applied to close the gap of the Young's
folds with four different ranges of pore sizes including 45–106 μm, modulus mismatch. The introduction of a β phase through alloying with
106–212 μm, 212–300 μm and 300–500 μm (Fig. 1). The Ti scaffolds β stabilizers [47–49] or sophisticated thermomechanical processes can
were seeded with osteosarcoma osteoblasts 143B and incubated for 12 reduce the Young's modulus of these metals. For example, Mo, tantalum
days for the evaluation of cell response in terms of cell attachment and (Ta), niobium (Nb) and zirconium (Zr) are effective β stabilizers and
proliferation. Result showed that small pores benefited for cell attach- can be used to alloy Ti, leading to a lower Young's modulus of these Ti
ment and large pores supported for cell proliferation (Fig. 2). The Ti alloys (∼42 GPa [50]). Nevertheless, natural bone still has lower
scaffold with the smallest pores (45–106 μm) exhibited excellent cell Young's modulus than that of Ti-based alloys. Another approach to
growth rate in the first three days, where the small pores provided more reducing the Young's modulus of metallic implant materials is to in-
surface area for cell attachment; however, the cell growth rate became troduce porosity into the metals [2,51–56]. According to the model of
slower than those on the scaffolds with larger pore sizes after 3 days of Gibson and Ashby [56], the most important structural characteristic of
cell culture. Ti scaffold with pore size 300–500 μm initially showed a a porous material that influences its mechanical response is its relative
lower cell growth rate in day one, but it tremendously increased to the density, ρ/ρs (the ratio of the density of the porous material ρ to that of
end of the culture time of 12 days [41]. Furthermore, Woodard et al. the solid material ρs). The relationships between the plastic collapse
[42] distinguished porosity into macro-porosity and micro-porosity strength, the elastic modulus, and the relative density can be approxi-
based on pore size. Macro-porosity is considered to involve a pore size mated by:
greater than 50 μm and micro-porosity involves a pore size less than σpl = 0.3(ρ/ρs)3/2σys (2)
20 μm. In scaffold design, having multiple porosities (a combination of
2
micro- and macro-porous structures) has shown better biocompatibility E = (ρ/ρs) Es (3)
than having only a macro-porous structure [43]. A scaffold with a
Eqs (2) and (3) indicate that the elastic modulus and plastic collapse
certain amount of micro-porosity increases the surface area for osteo-
strength both decrease with increasing porosity.
genic protein adsorption and osteoblast cell attachment, promoting
A mechanical evaluation of a biomedical scaffold could be compli-
rapid bone regeneration into the scaffold [43,44].
cated in the physiological environment, where the scaffold is under a
superimposed loading situation with load vectors coming from multiple
2.3. Effect of porosity on mechanical properties directions, rather than taking one simply measurement in uniaxial di-
rection [57]. In addition, most of the open cellular structures show
Mechanical properties are another important element that needs to

Fig. 1. Ti scaffolds with 70% porosity and different ranges of pore sizes [41].

58
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

Fig. 2. Cell viability of the porous Ti scaffolds with 70% porosity and different pore sizes after cell culture for 1, 3, 7 and 12 days [41].

anisotropic deformation behavior, thus the mechanical performance is AM theoretically simplifies the whole process and reduces unnecessary
affected by the pore orientation, arrangement, distribution, shape and cost [60]. AM has been developed over more than thirty years since the
size [58]. Yang et al. [58] studied the elastic behavior of gyroid cellular mid-1980s [61]. Nowadays, multiple AM approaches are available for
structure using compression tests and indicated that the loading di- different applications, based on the specific requirements of individual
rection significantly affects the Young's modulus of gyroid cellular objects. In general, fused deposition modelling (FDM), powder bed
structures. Xu et al. [50] investigated the microstructure and mechan- fusion (PBF), inkjet printing, stereolithography (SL), direct energy de-
ical properties of porous Ti architectures with different porosities, pore position (DED), and laminated object manufacturing (LOM) are the
sizes and distributions; unsurprisingly, the Young's modulus and yield main approaches of current AM.
stress were inversely proportional to the pore size and porosity. Their
results also indicated that porous Ti structures with a staggered pore 3.1. AM procedures
distribution exhibited a lower Young's modulus than those with a reg-
ular pore distribution [50]. Generally, AM technology operates from 3D object modelling via
computer aided design (CAD) to the fabrication of structures using a
2.4. Effect of porosity on biodegradability layer-by-layer 3D printing process. In the modelling phase, a variety of
software applications allows creating, modifying, and optimizing the
Biomaterials for temporary implant applications should exhibit an designed structures, which increases the productivity and improves the
appropriate degradation rate to support bone tissue regeneration. The quality of the final products. Once the objects have been designed by a
degradation rate of an implant depends on the corrosion resistance of CAD software application, the files have to be saved as standard tri-
the material, which is affected by the chemistry as well as the physical angulate language (STL) format for the subsequent printing purposes.
characteristics of the implant. For instance, the porosity of the material STL is a file format in CAD that supports 3D printing and computer
plays a vital role in affecting the corrosion rate. Through reducing the aided manufacturing (CAM). In the fabrication phase, the generated
porosity of the material, the corrosion resistance can be increased due STL files are input into the selected printing machines to build up the
to the decrease in the specific surface area, and vice versa. The corro- 3D models in a layer-by-layer process.
sion rate (r) can be calculated using the equation given by: In the AM of a medical implant, scanning of the host's own bone is a
necessary step, which has to be performed before the CAD modelling.
r (corrosion rate) (μg/day) = (M1-M2) / ti (μg/day) (4)
Magnetic resonance imaging (MRI), X-ray computed tomography (CT),
where M1 is the object mass before corrosion and M2 is the object mass and other scanning techniques can be used for the scanning to collect an
after corrosion, while ti is the immersion time [59]. accurate image data from each individual patient [62]. The scanned
Hou et al. [24] investigated the corrosion behavior of two groups of image data is manipulated and converted to 3D CAD models by using
Zn-based open-cellular scaffolds. The first group was pure Zn scaffolds specific biomaterial software applications such as Mimics® (Materialise,
with respective porosities of 68% and 75%, and the second group was Belgium) and Biobuild® (Anatomics, Australia) [63]. After the model
Zn-3wt%Cu (copper) scaffolds with the same respective porosities. The has been generated, the process is followed by the abovementioned
results showed that the corrosion rate was inversely proportional to the procedures for optimizing and 3D printing.
porosity over the whole time period for both groups of scaffolds.
3.2. Metallic AM techniques
3. An overview of additive manufacturing (AM)
As shown in Table 1, PBF, DED and LOM are capable for metallic
Recently, a novel fabrication technology, additive manufacturing materials fabrication. PBF uses metal powders as the material resources,
(AM), which is also known as rapid prototyping and three-dimensional whereas DED normally uses wires, LOM uses sheets [64]. DED, which is
(3D) printing, has been widely explored for both research and com- also called laser engineered net shaping (LENS), has accurate compo-
mercial purposes. This technology has shown great ability to manu- sition control and is capable of fabricating products with controlled
facture pieces from powders for diverse applications, thus eliminating microstructures thus excellent mechanical properties, specializing in
multiple manufacturing constraints and producing architecture with crack fillings and retrofitting of fabricated parts. The DED method
more complex geometry than with conventional technologies. presents lower accuracy of fabricated parts and lower surface quality,
Compared with traditional manufacturing processes, the AM process when compared with PBF. This method is used for large component
has the great advantage of process simplicity from the design phase to fabrication and broken parts repairing such as turbine blades repair in
installation. The whole AM process only takes few steps, which greatly aircraft. LOM can be used to print a wide range of materials, including
increases the productivity. In a traditional manufacturing process, a metal rolls. LOM can greatly reduce manufacturing time and tool cost.
large number of skilled operators are required for machining and Compared with other AM methods, LOM is the only approach for fab-
welding processes, resulting in high labor costs. In addition, the cost of ricating metallic structures at low temperature [65,66]. Apart from the
machines and maintenance expenses cannot be ignored, which makes above-mentioned advantages, LOM can be used for larger structure
up a notable proportion of the expense in the manufacturing process. fabrication; however, without post-processing the quality of the

59
 L. Yuan et al.

Table 1
Materials, general applications, product resolution, advantages and disadvantages, and build volumes of six different categories of additive manufacturing: fused deposition modelling, powder bed fusion, inkjet printing,
stereolithography, direct energy deposition and laminated object manufacturing. Reproduced and modified from Ref. [72].
Additive manufacturing

Category Material General applications Resolution (μm) Advantages Disadvantages Build volume References
(mm)

Fused deposition modelling (FDM) Continuous filaments of • Toys 50–200 • Simplicity • Low dimensional resolution X: < 900 [72,74–78]
thermoplastics (PLA, ABS, • Commercial samples • Fast processing time • Weakness of mechanical Y: < 600
PETG, nylon etc.) • Affordability properties (inner layer Z: < 900
distortion)
• Poor surface quality
• Limited materials
Powder bed fusion (Selective laser sintering SLS (polymers, metals and alloy • Biomedical fabrication 80–250 • Superior mechanical • Expensive X: 200–300 [72,77,79]
(SLS)/Selective laser melting (SLM)/ powders) • Electronics Aerospace properties • Time consuming process Y: 200–300
Electron beam melting (EBM)) SLM and EBM (specific metal etc. • High quality • Shrinkage or distortion Z: 200–350
materials) • Complex geometry
fabrication
Inkjet printing and contour crafting Ceramic structure for Inkjet: 5–200 process workability X: < 300 [72,77]

60
Concrete and soil for contour
• Advanced
tissue engineering Contour crafting:
• Time-effective
for construction
• Maintaining
resolution (coarseness) Y: < 200
crafting structure 25000–40000
• Used
(contour crafting)
• Low Z: < 200
• Building
fabrication
• Poor adhesion X: < 4000
Y: < 2000
Z: < 1000
Stereolithography (SL) Polymers (photopolymers) • Biomedical applications 10 •. High quality of printing • Expensive X: < 2100 [72,77,80]
Composites (can applied to • Commercial prototypes objects • Time-consuming process Y: < 700
Nano –composites) •. Fine resolution • Limited for large-volume Z: < 800
•. High fabrication accuracy production
•. Smooth finish
Direct energy deposition (DED)/Laser Metals and metal alloys •. Aerospace (turbine 250 • Excellent mechanical • Post-processing required X: 600–3000 [72,76,77,81]
engineered net shaping (LENS)/Direct Ceramics blade repair) properties • Low resolution Y: 500–3500
metal deposition DMD) Glass •. Retrofitting • Rapid cooling rates • Low surface quality Z: 350–5000
Polymers •. Repair of bespoke parts • Effective time and cost of • Producing less complex
•. Cladding repairs models (in comparison with
•. Biomedical applications • Multiple-axis deposition and powder bed fusion)
multiple materials at the
same time
• Controlled microstructure
Laminated object manufacturing (LOM) Polymers •. Paper manufacturing Based on laminate • Time-effective processing • Post process required X: 150–250 [72,77]
Metals •. Foundry industries thickness • Low cost • Limited materials Y: 200
Ceramics •. Electronics Smart • Poor surface quality and Z: 100–150
Paper structures accuracy
Bioactive Materials 4 (2019) 56–70
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

the raw materials, continuous filaments of thermoplastic polymers,

need to be heated up to a semi-liquid state and extruded onto a hor-
izontal XeY plane by a nozzle, building up layer by layer in the Z-axis
[72]. The FDM method has great advantages of simplicity, speed and
low cost. However, this method has the disadvantages of weak me-
chanical properties and poor surface quality. Inkjet printing can be used
to print ceramics with complex and advanced structures for bio-medical
scaffolds, but the issues of coarse resolution and lack of adhesion be-
tween layers remain to be addressed in the future. SL was introduced in
the late 1980s, and is one of the most powerful and versatile solid
freeform techniques to fabricate high-quality printings with fine re-
solution [73]. This technology has great potential for printing bio-
compatible and biodegradable devices in tissue engineering. However,
Fig. 3. Schematic of an SLM machine [84]. the major concerns of SL are the limitations in materials and expense.

3.4. SLS and SLM

product surface is poor, and it has certain limitations in manufacturing
complex shapes. The PBF technique can be used to produce 3D models
SLS can be used to produce novel structures and parts by solidifying
with good mechanical properties and fine resolution, printing complex
powders in a layer-by-layer process. The laser is required to provide
structures such as triply periodic minimal surface (TPMS) structures.
enough thermal energy for powder sintering, where the temperature of
This technique has been used in a variety of fields such as aerospace,
the powder should be raised above the melting point of metallic ma-
electronics and metal scaffolds in bone tissue engineering. PBF is one of
terials or the softening point of polymers. In some cases of metallic
the laser-based AM techniques, which builds up 3D objects by using
applications in SLS, binder materials (materials that have low melting
laser beams to scan a powder bed at a predefined speed and location.
points) are introduced to reduce the melting point, thus promoting the
The laser selectively fuses the powders at the surface, either completely
sintering process [72,79]. The fabrication of the porous poly-
melted as selective laser melting (SLM) or partially melted as selective
caprolactone scaffold is a good example of SLS, which has shown great
laser sintering (SLS). After a scanning process, the powder re-
improvement in reducing the stiffness of the scaffold to the range of
consolidates into solid form. Once a layer is completed, the build
300–400 KPa [82]. SLM is a similar fabrication process to SLS. How-
platform moves downwards in the z direction for a defined distance
ever, there are differences between the two techniques in relation to the
(thickness of the layer) to print another layer. This process follows a
type of raw materials and the bonding process between particles [83].
layer-by-layer sequence until the object is fully printed. SLS, SLM and
SLS can be used to fabricate different types of material, including
EBM are the sub-classifications of powder bed fusion techniques [67].
polymers, metals, and alloys, whereas SLM can be used to fabricate
Compared with other AM techniques, SLS, SLM and EBM have great
certain metal products such as Ti, stainless steel, cobalt-chromium
advantages in implant fabrication. These methods can be used to fab-
(CoeCr) alloys, and aluminum (Al) [79]. As previously mentioned, the
ricate porous orthopedic implants directly. The porous structures fa-
laser scanning in SLS involves the partial melting and resolidification of
cilitate both bone regeneration and ingrowth in load-bearing applica-
powder particles, whereas the powders are completely melted in the
tions in which high fracture toughness and mechanical strength are
laser scanning in SLM. A schematic diagram of DIMetal-100 SLM ma-
required [68]. In addition, SLS, SLM and EBM are capable of fabricating
chine is shown in Fig. 3, where the build volume is 100 × 100 × 120
metal structures with complex geometry, such as open cellular struc-
mm3 [84]. The SLM machine generates high energy laser beam through
tures [69–71]. Apart from the abovementioned techniques, there are
F-theta lens to fully melt the powder particles along the predefined
some other AM techniques not shown in Table 1, which are available
path. The melted powder particles will be rapidly solidified and form
for metal fabrication such as binder jetting. Similar to PBF, binder
the shape of the thin layer. Once the layer has been completed, the
jetting uses metal powders as the raw material and this technique is
building cylinder will go downward in a predefined thickness to start
capable of processing various metals and alloys including Al-based, Cu-
another layer fabrication [85]. As shown in Table 2, the layer thickness
based, Fe-based, Ni-based, and Co-based alloys. However, binder jet-
of SLM is in the range of 0.020–0.100 mm [86]. The printing process is
ting-built parts possess lower mechanical properties than SLM or EBM-
undertaken in controlled conditions and an inert gas such as argon (Ar)
built parts [64].
or nitrogen (N) is used to fill the build chamber to prevent oxidation
during the fabrication process [69,84].
3.3. Other AM techniques In SLM, fully melted powder particles can provide enhanced
bonding between particles and therefore improved the mechanical
FDM is the most straightforward method of 3D printing, in which properties of the products. SLM is capable of producing complex

Table 2
Features of SLM and EBM in comparison [86].
SLM EBM

Powder sources One or more fiber lasers of 200–1000 W High power Electron beam of 3000 W
Build chamber environment Argon or Nitrogen Vacuum/He bleed
Method of powder preheating Platform heating Preheat scanning
Powder preheating temperature (°C) 100–200 700–900
Maximum available build volume (mm) 500 × 350 × 300 350 × 380 (diameter × length)
Maximum build rate (cm3/h) 20–35 80
Layer Thickness (mm) 0.020–0.100 0.050–0.200
Melt pool size (mm) 0.1–0.5 0.2–1.2
Surface finish (Ra) 4–11 25–35
Geometric tolerance (mm) ± 0.05–0.1 ± 0.2
Minimum feature size 40–200 100

61
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

Fig. 4. SEM images of Tie6Ale4V gyroid lattice surfaces fabricated by SLM: (a) and (b) as-built, (c) and (d) after post treatments (heat treatment and sandblasting)
[88].

structures without constraint of geometry. This technique also allows capable of fabricating brittle metals. As the rapid cooling rate could
recycling of the un-melted powders [64,87]. However, the dis- trigger solidification cranks of brittle metals, EBM technique accom-
advantages of SLM are obvious, such as high initial cost and acute size modates metal powers using a hot powder bed generally heated up to
restrictions. It should be noted that brittle materials should be avoided ∼870 K, therefore reducing the temperature difference between the
for SLM fabrication because solidification cracks may occur. Rapid metal powders and the powder bed, leading to reduced cooling rate of
heating and cooling of brittle metal powder in SLM process will lead to the molten powders [64]. However, under such circumstances, the
residual stress, thus resulting in cracks in the as-built parts [64,86]. processing time of EBM could be longer than SLM in order to cool down
Additionally, SLM-built parts normally have rough surface (Fig. 4(a and the as-built parts to room temperature. In addition, EBM-built parts also
b)) [88] due to the attachment of a large amount of partially melted exhibit rough surface because of the attachment of a large amount of
particles from SLM process. These unexpected particles could have partially melted particles (Fig. 5(a)), which is similar to SLM [20]. The
adverse effect on the mechanical properties especially the fatigue EBM-built scaffold was blasted by ceramic beads to improve the surface
strength [20,89]. Thus, post treatments are required to improve the quality (Fig. 5(b)) [20]. Apart from those drawbacks, this technique still
quality of the surfaces. As shown in Fig. 4(c and d), the surface has some other critical deficiencies in regards to process stability, de-
roughness of strut was improved by post treatments (heat treatment lamination occurrence and low accuracy [96]. In order to improve the
and sandblasting), where most of partially melted particles were re- reliability of EBM production, further research is needed to address all
moved [88]. these issues.
A schematic of an Aram EBM machine and its processing chamber is
3.5. EBM shown in Fig. 6 [91,92]. The EBM machine has two main components, a
processing chamber and a control panel. A tungsten filament is fixed on
EBM is another advanced powder bed fusion-based rapid proto- the upper column under the electron beam gun, which emits electrons
typing process for the fabrication of metal products. EBM uses an by heating up to a temperature ranging from 1900 K to 2200 K, and the
electron beam to melt conductive metal powders in a layer-by-layer generated electron beam is accelerated to a kinetic energy of about
process. A paper describing the capability of electron beam in the 60 KeV. The electron beam is adjusted by different magnetic lenses (or
manufacturing of 3D objects was published in 1994 [90,91]. Since then, coils), including an astigmatism lens, focus lens, and deflection lens.
this technique has been well developed. A Swedish company called The first magnetic lens is used for astigmatism correction; the second
Arcam EBM manufactures and distributes their EBM machines for magnetic lens focuses the electron beam to the desired diameter from
commercial purposes and the first production of EBM machines was 0.1 to 1 mm; and the third magnetic lens deflects the focused beam to a
launched in 2002 [92]. Presently, several kinds of metallic powders are desired position on the powder bed. In the processing chamber, the
available at Arcam for EBM fabrication, including Tie6Ale4V, Grade 2 metal powders, supplied from powder hoppers, are scanned by the
Ti, CoeCr alloy, and Inconel super 718 [92]. Similar to other powder electron beam. Similar to SLM, the scanned metal powder is first melted
bed fusion techniques, EBM fabrication has addressed some of the completely and then solidified into a thin layer with a thickness in the
limitations of conventional tool-cutting methods and it can be used to range of 0.05–0.2 mm. As the predefined pattern has been set in the
fabricate complex structures and parts, including porous scaffolds with computer, the electron beam scans the powder at a preset speed and
specified stiffness [93]. In an early study in 2008, Heinl et al. [94] location. In order to prevent oxidation of metal powders, especially for
successfully fabricated a porous Tie6Ale4V structure by EBM and in- highly reactive metals like Ti, the entire process takes place in vacuum
dicated that their models had reduced stiffness and were suitable for [86]. The chamber pressure is 10−1 Pa and the electron beam gun is at
bone ingrowth. Moreover, EBM is also considered a cost-effective pro- a pressure of 10−3 Pa. In the processing chamber, a low-pressure inert
cess for the fabrication of customized orthopedic implants and instru- helium (He) gas (10−1 Pa) is introduced to cool down the parts after the
ments for biomedical devices [95]. Unlike SLM fabrication, EBM is printing process. The He gas also protects the powders from becoming

62
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

Fig. 5. SEM images of Tie6Ale4V gyroid lattices surfaces fabricated by EBM: (a) as-built, (b) after post treatment of ceramic blasting [20].

electrically charged [91,96–101] (see Table 2). used for processing metallic biomaterials for load-bearing implants
applications. In an early research by Krishna et al. [104], the LENS
3.6. Metallic biomaterials fabricated by AM for implant applications technique was used in the fabrication of porous CPeTi with low stiff-
ness for load-bearing implant applications in 2007. Wu et al. [105]
Metal-AM fabrication has been extensively explored in the last investigated the mechanical response of gradient porous Tie6Ale4V
decade and this technology has been successfully used in the biome- cylinders fabricated by EBM with 70%–90% porosity, which exhibited
dical field. 3D printed physical models can provide a detailed visuali- different elastic moduli close to those of natural bone. Ataee et al. [106]
zation of clinic cases before surgery, which is a practical approach for reported ultra-high strength in a gyroid porous CPeTi fabricated by
surgery planning and accurate diagnosis. Surgeons and medical doctors SLM, which exhibited an elastic modulus close to that of trabecular
can simulate surgery processes on 3D printed models to examine the bone. Surmeneva et al. [107] fabricated Tie6Ale4V lattice scaffolds via
outcomes of the surgery and find any potential surgical risk and failing EBM and achieved a set of mechanical properties matching those of
factors [62,102]. More importantly, AM technology can produce cus- trabecular bones. In addition, AM techniques have also been used for
tomized implants for bone replacement and fixation. Since the current the fabrication of dental implants. Ti and CoeCr based alloys are the
implant structure design becomes increasingly complex to meet dif- favorable metallic materials for fabricating dental parts due to their
ferent requirements (e.g., interconnected porous TPMS scaffold), it excellent mechanical performance and surface properties [102]. How-
could be challenging to manufacture such a metallic implant via con- ever, there is a serious concern about metallic materials for dental ap-
ventional fabrication methods such as casting, forging, milling, and plications because alloys may release toxic ions which may cause ad-
turning. This technology is also capable of fabricating complex porous verse tissue reactions as well as hypersensitivity in individuals,
structures with both micro and macro porosity. Due to the high soli- especially in the oral environment [102,108]. Xin et al. [108] in-
dification rate of AM techniques, AM-produced implants may exhibit vestigated the biological response of CoCr samples in artificial saliva
high strength [103]. solution using a seven-day immersion test, where the samples were
Metallic biomaterials such as Ti alloys are widespread used for load- fabricated either by SLM or traditional casting. They found that the
bearing bone substitutes due to their excellent mechanical, corrosion SLM-built parts showed significantly low Co-ion release compared with
and biological properties. LENS, EBM and SLM have been extensively that of cast specimen. Thus, the SLM-built part showed a significant

Fig. 6. (a) Schematic of an EBM machine and (b) its processing chamber [91,92].

63
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

higher cell proliferation rate than that of the cast specimen. The SLM-
built CoCr samples exhibited higher biocompatibility and lower biolo-
gical risks than their cast counterparts. Apart from the load-bearing and
dental applications, AM techniques also show great capability in other
biomedical applications; for instance, Jardini et al. [63] reconstructed a
Tie6Ale4V cranial implant by using Direct Metal Laser Sintering
(DMLS) to restore the structure integrity of the patient's skull and facial
aesthetics. Demir et al. [109] successfully produced cardiovascular
CoCr stents by SLM and suggested that SLM could be an alternative Fig. 8. Gyroid surface following mathematical equation (3): in order to gen-
process for microtube manufacturing and laser microcutting in stent erate a basic unit cell of a gyroid surface, the x, y and z spatial directions are in
fabrication. the interval length of 2π, where x, y, z = [-π, π] and a = 0 .

4. Triply periodic minimal surface (TPMS) structures 70% porosity for the designed objects, the α offset value could be |0.60|
for a gyroid and |0.42| for a Schwarz diamond [115] (see Fig. 7).
4.1. Overview A computer program called Mathmod [116] has been introduced for
TPMS structural analysis, and provides support for visualization and
TPMS describes a periodically infinite structure along three in- manipulation of different TPMS models in multiple dimensions by
dependent directions with zero mean curvature of the surface (the mathematical equations. In recent years, this software application has
concave and convex curvatures are symmetrical at all points) drawn attention for the building of models. Mathmod was initially
[110,111]. Porous architectures with TPMS topology are constructed by applied to generate the unit cells of the desired TPMS models in many
repeating elements with the minimum possible area (unit cells) studies [117–122], and the generated models can be imported into
[110,112]. TPMS is also defined by implicit functions and can be dis- other software applications for further model construction.
tinguished by curved surfaces even at the junction of struts. Each type Single unit cells of gyroid and diamond surfaces created by
of TPMS architecture can be expressed in a strict mathematical equa- Mathmod are shown in Fig. 8 and Fig. 9, respectively.
tion. Typical examples of implicit TPMS implicit functions are given by Previous studies investigated the mechanical properties of TPMS
Ref. [111]: architectures [20,31,106,123,125] in which powder bed fusion fabri-
Gyroid surface: F (x , y, z ) = cos (x )⋅sin (y ) + cos (y )⋅sin (z ) + cos (z ) cation methods (SLS, SLM and EBM) were employed to build up me-
tallic TPMS structures, especially for Ti and its alloys (see Table 3).
⋅sin (x ) + a (5) These AM methods address the challenges of conventional manu-
facturing processes, allowing the fabrication of complex, open-cellular
Schwarz diamond: F (x , y, z ) = sin (x ) sin (y )⋅sin (z ) + sin (x )⋅cos (y ) TPMS architectures. TPMS porous structures and scaffolds possess great
⋅cos (z ) + cos (x )⋅sin (y )⋅cos (z ) + cos (x ) bio-mimicry features and excellent mechanical property, which im-
⋅cos (y )⋅sin (z ) + a (6) proves implant feasibility and reliability. According to Rajagopalan
et al. [126], the ideal biomorphic geometry would be a structure with
Neovius surface: F (x , y, z ) = 3⋅(cos (x ) + cos (y ) + cos (z )) + 4 surfaces that are continuous through space and separated into two not-
necessarily-equal sub-spaces by a non-intersecting, two-sided surface.
⋅(cos (x )⋅cos (y )⋅cos (z )) + a (7) TPMS geometries exhibit great biomorphic features with smooth con-
tinuous surfaces along three independent directions and the surfaces
D − prime surface: F (x , y, z ) = 0.5⋅(sin (x )⋅sin (y )⋅sin (z ) + cos (x ) divide the structure into two non-intersecting spaces. Compared to the
⋅cos (y )⋅cos (z )) − 0.5⋅(cos (2x ) straight edges or sharp turns of geometrically primitive's shapes such as
⋅cos (2y )þcos (2y )⋅ cos (2z ) + cos (2z )⋅cos (2x )) cylinders and cubes, TPMS structures can provide a better biomorphic
environment for cell activities such as cell attachment, penetration,
+a (8)
migration and proliferation [123,126]. In addition, TPMS structures
where the x, y, and z are in the Cartesian Coordinates [113]. The α were also reported to exhibit good fatigue properties because of the
variant is the offset value, which can be used to determinate the de- curvature of the struts; at the nodal points, the stress concentrations
signed solid fraction [31]. The offset value can be both positive and caused by defects were eliminated [117,119,127]. During the fabrica-
negative. If the offset value is positive, the gyroid surface build up in tion process, a TMPS structure has better manufacturability than a
positive region and vice versa [114]. For instance, in order to maintain structure with sharp turns or straight-edged pores and struts,

Fig. 7. Gyroid unit cell with ± 0.6 offset.

64
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

ingrowth [39,94]. Sheet solid gyroid structure displays unconnected

pores thus is not preferred by researchers in the field of tissue en-
gineering. Compared with sheet solid gyroid structure, network solid
gyroid structure also provides excellent mechanical properties. More
importantly, the pores are interconnected and their volume fraction is
controllable. This makes the structure more reliable and practical for
orthopedic implant applications (see Fig. 11).
The abovementioned equation (5) can be extended as:
Fig. 9. Diamond surface following mathematical equation (4): in order to 2π 2π 2π 2π 2π
F (x , y, z ) = cos ⎛ x ⎞⋅ sin ⎛ y ⎞ + cos ⎛ y ⎞⋅sin ⎛ z ⎞ + cos ⎛ z ⎞
generate a basic unit cell of a diamond surface, the x, y and z spatial directions ⎝ L ⎠ ⎝ L ⎠ ⎝ L ⎠ ⎝ L ⎠ ⎝ L ⎠
are in the interval length of 2π, where x, y, z = [-π, π] and a = 0 . 2π
⋅sin ⎛ x ⎞ + a
⎝ L ⎠ (9)
eliminating the issue of thermal deformation caused by long overhangs
where L is a variable determining the size of unit cell's edge length.
[128].
Offset a determines the volume fractions of the unit cell, enabling the
adjustment of porosities between 50% and 90% [114,134]. The volume
4.2. Popular TPMS structure – gyroid fraction of a gyroid unit cell is 50% when a = 0. With an increase in
the absolute offset value, the relative density of a gyroid structure de-
The most popular TPMS architectural shape for biomedical appli- creases and the porosity increases. Fig. 12(a–f) show the gyroid surfaces
cations is the gyroid minimal structure, discovered by Schoen in 1970 and network based on gyroid unit cell with different α values. In
[129,130]. Researchers have experimentally shown that the gyroid Fig. 12(c), the porosity of the gyroid structure is 70% when a = −0.6.
architecture is suitable for biomorphic scaffold design in tissue en- According to Dawei et al. [114], when the absolute value of a = 1.41,
gineering [112]. Recently, extensive research has been carried out on the gyroid structure exhibits the maximum porosity of 90% and be-
analysis of the mechanical properties of open-cellular gyroid structures. comes a ‘pinch-off’ phase (Fig. 12(e)), which causes geometric dis-
From the morphological perspective, TPMS architectures are generally continuity problems. This gyroid structure becomes fragile and losses
categorized into two types, minimal surface network solids and its mechanical properties, and the manufacturability of such structures
minimal surface sheet solids, which can also be applied to gyroid sur- also decreases. When the absolute value of a > 1.41, the struts of the
face structures [131–133]. Both structures are established from gyroid gyroid structure collapse into eight small parts and all parts are dis-
surfaces and are distinguished by the number of void spaces. A net- connected (Fig. 12(f-g)). In addition, the absolute a value can not be
work-based gyroid structure is a solid structure with a single void do- great than 1.5 as the gyroid will disappear [114].
main, whereas a sheet-based gyroid structure has two partitioned, un-
connected void spaces. Kapfer et al. [131] demonstrated that the sheet- 4.3. Normal gyroids and deformed gyroids
based gyroid structure has higher stiffness than the network-based gy-
roid structure at the same porosity of the same material. Similarly, Al- In order to improve the mechanical performance of the gyroid
Ketan et al. [133] investigated the mechanical properties of a wide structures, the gyroid pore shape can be modified to be the deformed
range of structures, including strut-based, skeletal-TPMS and sheet- gyroids. The normal gyroid architecture is associate with spherical
TPMS porous structures, and concluded that sheet-TPMS structures pores, where the angle between the strut and the axial direction is 45°.
have superior mechanical properties in terms of stress and strain re- A deformed gyroid exhibits ellipsoidal-shaped pores with the variable
sponses (Young's modulus). Sheet-based gyroid structures also have radius in the direction of the longitudinal axis. Fig. 13 shows the
relatively higher Young's modulus, peak stress, and toughness in com- schematic of a normal pore and a deformed pore. Both gyroid structures
parison with skeletal gyroid structures (see Fig. 10). have shown a good strength-to-weight ratio for a certain angle of strut
Although sheet solid gyroid structure shows good mechanical per- orientation and specific strength and stiffness [135,136]. Yanez et al.
formance, unconnected porous structure could be a serious dis- [31,135] investigated the mechanical properties of a normal gyroid
advantage. As mentioned earlier, good pore interconnectivity is essen- scaffold and a severally deformed gyroid scaffold with different angels
tial for scaffold design because the interconnected space supports body (19°, 21.5°, 26°, 35°, 55°, 64° and 68.5°) in terms of compression tests,
fluid transport, facilitates vascularization, cells activities and tissue torsion tests and finite element analysis and reported that both the

Table 3
Mechanical properties and porosity of SLM/EBM fabricated Tie6Ale4V/CP-Ti TPMS architectures and human cortical and trabecular bone.
Structure Porosity (Volume Material Unit cell Elastic modulus (GPa) Pore size (mm) Methods Yield strength (MPa) References
(Scaffolds) fraction %) size (mm)

Gyroid 80–95 Tie6Ale4V 3–7 0.13 ± 0.02–1.25 ± 0.04 0.56–1.6 SLM 6.501± 0.62–81.30± 2.60 [123]
Gyroid 5–10 Tie6Ale4V 17.45–19.14 (Predicted) – 1342-1581 (Predicted) [123]
Diamond 80–95 Tie6Ale4V 3–7 0.12 ± 0.03–1.25 ± 0.07 0.48–1.45 SLM 4.662 ± 0.13–69.21 ± 4.22 [123]
Diamond 5–10 Tie6Ale4V 15.73–17.19 (predicted) 1403-1559 (Predicted) [123]
Gyroid – Tie6Ale4V 2.5–4 0.5912–0.7 – EBM 1.69–13.19 [31]
Gyroid 82 Tie6Ale4V 2 0.637–1.084 1.85–0.6 EBM 13.18–19.20 [20]
Gyroid 84.5 Tie6Ale4V 2.5 0.842–1.060 1.24–0.88 EBM 15.53–17.35 [20]
Gyroid 85 Tie6Ale4V 3 0.839–0.824 1.5–1.47 EBM 14.05–15 [20]
Gyroid 68.7 CPeTi (grade 2 2.017–2.676 1.24 ± 0.1 SLM 51.6–54.5 [106]
1)
Gyroid 73.3 CPeTi (grade 2.5 2.107–2.170 1.66 ± 0.1 SLM 44.9–56.5 [106]
1)
Gyroid 72.4 CPeTi (grade 3 1.465–2.302 1.91 ± 0.1 SLM 49.0–53.3 [106]
1)
Cortical bone 5–10 Human bones 7.7–21.8 103–222 [20,106,123,124]
Trabecular bone 50–90 Human bones 0.0224–0.132.32 0.8–11.6 [ [21,106,123,124]]

65
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

Fig. 10. 3D CAD gyroid unit cells: (a) 3 mm sheet solid gyroid unit cell with 0.3 mm offset thickness and (b) 3 mm network solid gyroid unit cell at 50% volume
fraction.

Fig. 11. A block of a 3D CAD gyroid scaffold in different views (constituted by 3 mm network solid gyroid unit cell).

elastic modulus and compressive strength of the gyroid structures were patients from biocompatibility and mechanical risks. On the one
reversely proportional to the strut angle at the axial direction. As the hand, implants need to be accepted by the human body, ensuring
pores of normal gyroid exhibit a spherical shape, the structure possesses cells live in a safe and supportive environment for their activities.
higher homogeneity in mechanical performance than that of the de- On the other hand, mechanical damage or failure caused by stress-
formed gyroid. An optimization of gyroid structure to support different shielding needs to be avoided. An implant needs to have an ap-
types of loads at different directions might be a suitable solution for the propriate elastic modulus to match that of natural bone.
reconstruction of bone defects in the human body [31]. (2) A well-designed porous structure is beneficial for cell activities such
as attachment and proliferation, which facilitate new bone forma-
5. Conclusions tion and regeneration. In addition, porosity is a parameter that can
be used to adjust the mechanical properties of the designed struc-
In summary, this paper has reviewed the current technologies for tures, in terms of both strength and Young's modulus. In the case of
open-cellular structural design for metal implant applications. The temporary metal implants, porosity plays a vital role in determining
fundamental requirements of metal implants, porosity, fabrication the corrosion rate, due to the change in the surface area.
methods and TPMS have been discussed. The main conclusions are as (3) One of the advanced AM methods, the powder bed fusion tech-
follows: nique, has great potential for metal implant fabrication, and can be
used to manufacture high-quality metallic models. The sub-classes
(1) Metallic orthopedic implants must satisfy certain criteria to protect of the powder bed fusion method, SLS, SLM and EBM, have been

66
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

Fig. 12. Gyroid surfaces and network-based on gyroid unit cell with different offset (α) values: (a) a 3 mm network-based gyroid structure in an 3 × 3 × 3 mm cubic;
(b-1) gyroid surface without offset, (b-2) network-based gyroid unit cell without offset, (c-1) gyroid surface with offset = −0.6, (c-2) network-based gyroid unit cell
with offset = −0.6, (d-1) gyroid surface with offset = −1.31, (d-2) network-based gyroid with offset = −1.31, (e) gyroid surface with offset = −1.41, (f) gyroid
surface with offset = −1.42, (g) gyroid surface with offset = −1.49.

Fig. 13. Schematic of a normal pore and a deformed pore.

applied to implant fabrication in biomedical and biomaterial stu- the Australian Research Council through the Discovery Projects
dies. DP170102557 and DP180100762.
(4) Implant structures with TPMS are drawing researching attention.
These novel structures, such as gyroid structure, exhibit great bio- References
mimicry features to support cell activity and also provide superior
mechanical properties. [1] E.B. Goudie, C. Robinson, P. Walmsley, I. Brenkel, Changing trends in total knee
replacement, Eur. J. Orthop. Surg. Traumatol. 27 (2017) 539–544 http://doi.org/
10.1007/s00590-017-1934-8.
[2] E. Losina, J.N. Katz, Total knee arthroplasty on the rise in younger patients: are we
Acknowledgements sure that past performance will guarantee future success, Arthritis Rheum. 64
(2012) 339–341 http://doi.org/10.1002/art.33371.
[3] K. Prasad, O. Bazaka, M. Chua, M. Rochford, L. Fedrick, J. Spoor, R. Symes,
The authors acknowledge the financial support for this research by

67
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

M. Tieppo, C. Collins, A. Cao, D. Markwell, K. Ostrikov, K. Bazaka, Metallic bio- [33] F.S. Kaplan, W.C. Hayes, T.M. Keaveny, A. Boskey, T.A. Einhorn, J.P. Iannotti,
materials: current challenges and opportunities, Materials 10 (2017) 884 https:// Form and Function of Bone, Orthopaedic Basic Science, American Academy of
doi.org/10.3390/ma10080884. Orthopaedic Surgeons, 1994, pp. 128–184.
[4] L.C. Palmer, C.J. Newcomb, S.R. Kaltz, E.D. Spoerke, S.I. Stupp, Biomimetic sys- [34] T.M. Keaveny, E.F. Morgan, G.L. Niebur, O.C. Yeh, Biomechanics of trabecular
tems for hydroxyapatite mineralization inspired by bone and enamel, Chem. Rev. bone, Annu. Rev. Biomed. Eng. 3 (2001) 307–333 https://www.annualreviews.
108 (2008) 4754–4783 https://pubs.acs.org/doi/abs/10.1021/cr8004422. org/doi/abs/10.1146/annurev.bioeng.3.1.307.
[5] Therapeutic Goods Administration, Australian Regulatory Guidelines for Medical [35] G.A.P. Renders, L. Mulder, L. J van Ruijven, T.M.G.J. van Eijden, Porosity of
Devices, (2011) https://www.tga.gov.au/sites/default/files/devices-argmd-01.pdf human mandibular condylar bone, J. Anat. 3 (2010) 239–248 https://doi.org/10.
, Accessed date: 8 June 2018. 1111/j.1469-7580.2007.00693.x.
[6] Statutory Rules No. 236, Therapeutic Goods (Medical Devices) Regulations 2002, [36] J. Wieding, T. Lindner, P. Bergschmidt, R. Bader, Biomechanical stability of novel
(2002) https://www.legislation.gov.au/Details/F2018C00490 , Accessed date: 8 mechanically adapted open-porous titanium scaffolds in metatarsal bone defects of
June 2018. sheep, Biomaterials 46 (2015) 35–47 https://doi.org/10.1016/j.biomaterials.
[7] Science Learning Hub, Toxicity, (2009) https://www.sciencelearn.org.nz/ 2014.12.010.
resources/1540-toxicity , Accessed date: 9 June 2018. [37] J. Wieding, A. Jonitz, R. Bader, The effect of structural design on mechanical
[8] Department of health, What You Know Can Help You - an Introduction to Toxic properties and cellular response of additive manufactured titanium scaffolds,
Substances, (2013) https://www.health.ny.gov/environmental/chemicals/toxic_ Materials 5 (2012) 1336–1347 https://doi.org/10.3390/ma5081336.
substances.htm , Accessed date: 8 June 2018. [38] Y. Kuboki, H. Takita, D. Kobayashi, E. Tsuruga, M. Inoue, M. Murata, et al., BMP-
[9] M. Jaishankar, T. Tseten, N. Anbalagan, B.B. Mathew, K.N. Beeregowda, Toxicity, induced osteogenesis on the surface of hydroxyapatite with geometrically feasible
mechanism and health effects of some heavy metals, Interdiscipl. Toxicol. 7 (2014) and nonfeasible structures: topology of osteogenesis, J. Biomed. Mater. Res. 39
60–72 https://doi.org/10.2478/intox-2014-0009. (1998) 190–199 https://doi.org/10.1002/(SICI)1097-4636(199802)
[10] Q. Chen, G.A. Thouas, Metallic implant biomaterials, Mater. Sci. Eng. R Rep. 87 39:2<190::AID-JBM4>3.0.CO;2-K.
(2015) 1–57 https://doi.org/10.1016/j.mser.2014.10.001. [39] Q. Chen, O. Bretcanu, A.R. Boccaccini, Inorganic and composite bioactive scaffolds
[11] L.M. Plum, L. Rink, H. Haase, The essential toxin: impact of zinc on human health, for bone tissue engineering, in: P.K. Chu, X. Liu (Eds.), Biomaterials Fabrication
Int. J. Environ. Res. Publ. Health 7 (2010) 1342–1365 https://doi.org/10.3390/ and Processing Handbook, CRC Press, 2008.
ijerph7041342. [40] A. Vahidgolpayegani, C. Wen, P. Hodgson, Y. Li, 2 - production methods and
[12] K.S. Egorova, V.P. Ananikov, Toxicity of metal compounds: knowledge and myths, characterization of porous Mg and Mg alloys for biomedical applications, Metallic
Organometallics 36 (2017) 4071–4090 https://pubs.acs.org/doi/full/10.1021/ Foam Bone, 2017, pp. 25–82 https://doi.org/10.1016/B978-0-08-101289-5.
acs.organomet.7b00605. 00002-0.
[13] D.F. Williams, There is no such thing as a biocompatible material, Biomaterials 35 [41] C. Torres-Sanchez, F.R.A. Al Mushref, M. Norrito, K. Yendall, Y. Liu, P.P. Conway,
(2014) 10009–10014 https://doi.org/10.1016/j.biomaterials.2014.08.035. The effect of pore size and porosity on mechanical properties and biological re-
[14] F.J. O'Brien, Biomaterials & scaffolds for tissue engineering, Mater. Today 14 sponse of porous titanium scaffolds, Mater. Sci. Eng. C 77 (2017) 219–228 https://
(2011) 88–95 https://doi.org/10.1016/S1369-7021(11)70058-X. doi.org/10.1016/j.msec.2017.03.249.
[15] N. Little, B. Rogers, M. Flannery, Bone formation, remodeling and healing, Surgery [42] J.R. Woodard, A.J. Hilldore, S.K. Lan, C.J. Park, A.W. Morgan, J.A.C. Eurell,
29 (2011) 141–145 https://doi.org/10.1016/j.mpsur.2011.01.002. S.G. Clark, M.B. Wheeler, R.D. Jamison, A.J. Wagoner Johnson, The mechanical
[16] C.S. Terrier, P. Gasque, Bone responses in health and infectious diseases: a focus properties and osteoconductivity of hydroxyapatite bone scaffolds with multi-scale
on osteoblasts, J. Infect. 75 (2017) 281–292 https://doi.org/10.1016/j.jinf.2017. porosity, Biomaterials 28 (2007) 45–54 https://doi.org/10.1016/j.biomaterials.
07.007. 2006.08.021.
[17] K.L. Bundy, Biomaterials and the Chemical Environment of the Body, Joint [43] S. Bose, M. Roy, A. Bandyopadhyay, Recent advances in bone tissue engineering
Replacement Technology, (2008), pp. 56–80 https://doi.org/10.1533/ scaffolds, Trends Biotechnol. 30 (2012) 546–554 https://doi.org/10.1016/j.
9781845694807.1.56. tibtech.2012.07.005.
[18] L.A.G. Armas, J.M. Lappe, R.P. Heaney, Chapter 19 - calcium, bone strength and [44] K.A. Hing, B. Annaz, S. Saeed, P.A. Revell, T. Buckland, Microporosity enhances
fractures, Osteopor. Men (2010) 235–241 https://doi.org/10.1016/B978-0-12- bioactivity of synthetic bone graft substitutes, J. Mater. Sci. Mater. Med. 16 (2005)
374602-3.00019-5. 467–475 https://doi.org/10.1007/s10856-005-6988-1.
[19] A. Biesiekierski, J. Wang, M.A. Gepreel, C. Wen, A new look at biomedical Ti- [45] Q. Ran, W. Yang, Y. Hu, X. Shen, Y. Yu, Y. Xiang, K. Cai, Osteogenesis of 3D
based shape memory alloys, Acta Biomater. 8 (2012) 1661–1669 https://doi.org/ printed porous Ti6Al4V implants with different pore sizes, Journal of the
10.1016/j.actbio.2012.01.018. Mechanical Behavior of Biomedical Materials 84 (2018) 1–11 https://doi.org/10.
[20] A. Ataee, Y. Li, D. Fraser, G. Song, C. Wen, Anisotropic Ti-6Al-4V gyroid scaffolds 1016/j.jmbbm.2018.04.010.
manufactured by electron beam melting (EBM) for bone implant applications, [46] A.J. van den Bogert, Analysis and simulation of mechanical loads on the human
Mater. Des. 137 (2018) 345–354 https://doi.org/10.1016/j.matdes.2017.10.040. musculoskeletal system: a methodological overview, Exerc. Sport Sci. Rev. 22
[21] G. Poumarat, P. Squire, Comparison of mechanical properties of human, bovine (1994) 23–51.
bone and a new processed bone xenograft, Biomaterials 14 (1993) 337–340 [47] S. Ozan, J. Lin, Y. Li, R. Ipek, C. Wen, Development of Ti-Nb-Zr alloys with high
https://doi.org/10.1016/0142-9612(93)90051-3. elastic admissible strain for temporary orthopedic devices, Acta Biomater. 20
[22] M. Moravej, D. Mantovani, Biodegradable metals for cardiovascular stent appli- (2015) 176–187 https://doi.org/10.1016/j.actbio.2015.03.023.
cation: interests and new opportunities, Int. J. Mol. Sci. 12 (2011) 4250–4270 [48] A. Biesiekierski, J. Lin, K. Munir, S. Ozan, Y. Li, C. Wen, Extraordinary high
https://doi.org/10.3390/ijms12074250. strength Ti-Zr-Ta alloys through nanoscaled, dual-cubic spinodal reinforcement,
[23] P. Wen, L. Jauer, M. Voshage, Y. Chen, R. Poprawe, J.H. Schleifenbaum, Acta Biomater. 53 (2017) 549–558 https://doi.org/10.1038/s41598-018-
Densification behavior of pure Zn metal parts produced by selective laser melting 24155-y.
for manufacturing biodegradable implants, J. Mater. Process. Technol. 258 (2018) [49] A. Biesiekierski, J. Lin, Y. Li, D. Ping, Y. Yamabe-Mitarai, C. Wen, Impact of ru-
128–137 https://doi.org/10.1016/j.jmatprotec.2018.03.007. thenium on mechanical properties, biological response and thermal processing of
[24] Y. Hou, G. Jia, R. Yue, C. Chen, J. Pei, H. Zhang, H. Huang, M. Xiong, G. Yuan, ß-type Ti-Nb-Ru alloys, Acta Biomater. 48 (2017) 461–467 https://doi.org/10.
Synthesis of biodegradable Zn-based scaffolds using NaCl templates: relationship 1016/j.actbio.2016.09.012.
between porosity, compressive properties and degradation behavior, Mater. Char. [50] G. Xu, H. Kou, X. Liu, F. Li, J. Li, L. Zhou, Microstructure and mechanical prop-
137 (2018) 162–169 https://doi.org/10.1016/j.matchar.2018.01.033. erties of porous titanium based on controlling Young's modulus, Rare Metal Mater.
[25] G. Katarivas Levy, J. Goldman, E. Aghion, The prospects of zinc as a structural Eng. 46 (2017) 2041–2048 https://doi.org/10.1016/S1875-5372(17)30176-5.
material for biodegradable implants—a review paper, Metals 7 (2017) 402 [51] M. Fousová, D. Vojtěch, J. Kubásek, E. Jablonská, J. Fojt, Promising characteristics
https://doi.org/10.3390/met7100402. of gradient porosity Ti-6Al-4V alloy prepared by SLM process, Journal of the
[26] H. Li, Y. Zheng, L. Qin, Progress of biodegradable metals, Prog. Nat. Sci.: Mater. Mechanical Behavior of Biomedical Materials 69 (2017) 368–376 https://doi.org/
Int. 24 (2014) 414–422 https://doi.org/10.1016/j.pnsc.2014.08.014. 10.1016/j.jmbbm.2017.01.043.
[27] Y.F. Zheng, X.N. Gu, F. Witte, Biodegradable metals, Mater. Sci. Eng. R Rep. 77 [52] C. Wen, M. Mabuchi, Y. Yamada, K. Shimojima, Y. Chino, T. Asahina, Processing of
(2014) 1–34 https://doi.org/10.1016/j.mser.2014.01.001. biocompatible porous Ti and Mg, Scripta Mater. 45 (2001) 1147–1153.
[28] P.K. Bowen, J. Drelich, J. Goldman, Zn exhibits ideal physiological corrosion be- [53] C. Wen, Y. Yamada, K. Shimojima, Y. Chino, H. Hosokawa, M. Mabuchi, Novel
havior for bioabsorbable stents, Adv. Mater. 25 (2013) 2577–2582 https://doi. titanium foam for bone tissue engineering, J. Mater. Res. 17 (2002) 2633–2639.
org/10.1002/adma.201300226. [54] C. Wen, Y. Yamada, K. Shimojima, Y. Chino, T. Asahina, M. Mabuchi, Processing
[29] Y. Li, H. Jahr, K. Lietaert, P. Pavanram, A. Yilmaz, L.I. Fockaert, M.A. Leeflang, and mechanical properties of autogenous titanium implant materials, J. Mater. Sci.
B. Pouran, Y. Gonzalez-Garcia, H. Weinans, J.M.C. Mol, J. Zhou, A.A. Zadpoor, Mater. Med. 13 (2002) 397–401.
Additively manufactured biodegradable porous iron, Acta Biomater. 77 (2018) [55] C. Wen, Y. Yamada, K. Shimojima, Y. Chino, H. Hosokawa, M. Mabuchi,
380–393 https://doi.org/10.1016/j.actbio.2018.07.011. Compressibility of porous magnesium foam: dependency on porosity and pore
[30] A. Carlos, Leon y Leon, New perspectives in mercury porosimetry, Adv. Colloid size, Mater. Lett. 58 (2004) 357–360.
Interface Sci. 76–77 (1998) 341–372. [56] L.G. Gibson, M.F. Ashby, Cellular Solids: Structure and Properties, second ed.,
[31] A. Yánez, A. Cuadrado, O. Martel, H. Afonso, D. Monopoli, Gyroid porous titanium Cambridge University Press, Cambridge, U.K, 1997, p. 429.
structures: a versatile solution to be used as scaffolds in bone defect reconstruc- [57] V. Weißmann, R. Bader, H. Hansmann, N. Laufer, Influence of the structural or-
tion, Mater. Des. 140 (2018) 21–29 https://doi.org/10.1016/j.matdes.2017.11. ientation on the mechanical properties of selective laser melted Ti6Al4V open-
050. porous scaffolds, Mater. Des. 95 (2016) 188–197 https://doi.org/10.1016/j.
[32] V. Karageorgiou, D. Kaplan, Porosity of 3D biomaterial scaffolds and osteogenesis, matdes.2016.01.095.
Biomaterials 26 (2005) 5474–5491 https://doi.org/10.1016/j.biomaterials.2005. [58] L. Yang, C. Yan, H. Fan, Z. Li, C. Cai, P. Chen, Y. Shi, S. Yang, Investigation on the
02.002. orientation dependence of elastic response in Gyroid cellular structures, Journal of

68
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

the Mechanical Behavior of Biomedical Materials 90 (2019) 73–85 https://doi. [85] H. Shipley, D. McDonnell, M. Culleton, R. Coull, R. Lupoi, G. O'Donnell,
org/10.1016/j.jmbbm.2018.09.042. D. Trimble, Optimisation of process parameters to address fundamental challenges
[59] J. Kubásek, D. Vojtěch, Zn-based alloys as an alternative biodegradable material, during selective laser melting of Ti-6Al-4V: a review, Int. J. Mach. Tool Manufact.
Metals 5 (2012) 1355–1361. 128 (2018) 1–20 https://doi.org/10.1016/j.ijmachtools.2018.01.003.
[60] B. Redwood, The advantages of 3D printing, 3D hubs https://www.3dhubs.com/ [86] V. Bhavar, P. Kattire, V. Patil, S. Khot, K. Gujar, R. Singh, A review on powder bed
knowledge-base/advantages-3d-printing , Accessed date: 4 June 2018. fusion technology of metal additive manufacturing, 4th Int. Conf. Exhib. Addit.
[61] D. Tang, R.S. Tare, L.Y. Yang, D.F. Williams, K.L. Ou, R.O.C. Oreffo, Biofabrication Manuf. Technol. Banglore, India, 2014.
of bone tissue: approaches, challenges and translation for bone regeneration, [87] L.C. Ardila, F. Garciandia, J.B. González-Díaz, P. Álvarez, A. Echeverria,
Biomaterials 83 (2016) 363–382 https://doi.org/10.1016/j.biomaterials.2016.01. M.M. Petite, R. Deffley, J. Ochoa, Effect of IN718 recycled powder reuse on
024. properties of parts manufactured by means of selective laser melting, Phys.
[62] M. Javaid, A. Haleem, Additive manufacturing applications in orthopaedics: a Procedia 56 (2014) 99–107 https://doi.org/10.1016/j.phpro.2014.08.152.
review, J. Clin. Orthopaed Trauma 9 (2018) 202–206 https://doi.org/10.1016/j. [88] C. Yan, L. Hao, A. Hussein, Q. Wei, Y. Shi, Microstructural and surface mod-
jcot.2018.04.008. ifications and hydroxyapatite coating of Ti-6Al-4V triply periodic minimal surface
[63] A.L. Jardini, M.A. Larosa, R.M. Filho, C.A.C. Zavaglia, L.F. Bernardes, lattices fabricated by selective laser melting, Mater. Sci. Eng. C 75 (2017)
C.S. Lambert, D.R. Calderoni, P. Kharmandayan, Cranial reconstruction: 3D bio- 1515–1524 https://doi.org/10.1016/j.msec.2017.03.066.
model and custom-built implant created using additive manufacturing, J. Cranio- [89] E. Sallica-Leva, A.L. Jardini, J.B. Fogagnolo, Microstructure and mechanical be-
Maxillofacial Surg. 42 (2014) 1877–1884 https://doi.org/10.1016/j.jcms.2014. havior of porous Ti–6Al–4V parts obtained by selective laser melting, Journal of
07.006. the Mechanical Behavior of Biomedical Materials 26 (2013) 98–108 https://doi.
[64] P.K. Gokuldoss, S. Kolla, J. Eckert, Additive manufacturing processes: selective org/10.1016/j.jmbbm.2013.05.011.
laser melting, electron beam melting and binder jetting-selection guidelines, [90] R. Larson, Method and Device for Producing Three-dimensional Bodies, Google
Materials 10 (2017) 672 https://doi:10.3390/ma10060672. Patents, 1994.
[65] R. Hahnlen, M.J. Dapino, NiTi–Al interface strength in ultrasonic additive man- [91] A. Ataee, Y. Li, G. Song, C. Wen, Metal Scaffolds Processed by Electron Beam
ufacturing composites, Compos. B Eng. 59 (2014) 101–108. Melting for Biomedical Applications, Metallic Foam Bone, (2017), pp. 83–110
[66] A. Hehr, M.J. Dapino, Interfacial shear strength estimates of NiTi-Al matrix https://doi.org/10.1016/B978-0-08-101289-5.00003-2.
composites fabricated via ultrasonic additive manufacturing, Compos. B Eng. 77 [92] Arcam EBM, Electron Beam Melting– in the Forefront of Additive Manufacturing
(2015) 199–208 https://doi.org/10.1016/j.compositesb.2015.03.005. 2018: http://www.arcam.com/technology/electron-beam-melting/. (Accessed 7
[67] Additive Manufacturing Research Group, About Additive Manufacturing Powder July 2018).
Bed Fusion. http://www.lboro.ac.uk/research/amrg/about/ [93] L.E. Murr, S.M. Gaytan, F. Medina, H. Lopez, E. Martinez, B.I. Machado,
the7categoriesofadditivemanufacturing/powderbedfusion/, (Accessed 3 July D.H. Hernandez, L. Martinez, M.I. Lopez, R.B. Wicker, J. Bracke, Next- generation
2018). biomedical implants using additive manufacturing of complex cellular and func-
[68] H.N. Chia, B.M. Wu, Recent advances in 3D printing of biomaterials, J. Biol. Eng. tional mesh arrays, Philos. Trans. R Soc. A Math. Phys. Eng. Sci. 368 (2010)
(2015), https://doi.org/10.1186/s13036-015-0001-4. 1999–2032 https://doi.org/10.1098/rsta.2010.0010.
[69] S. Sun, M. Brandt, M. Easton, Powder Bed Fusion Processes: an Overview, [94] P. Heinl, L. Müller, C. Körner, R.F. Singer, F.A. Müller, Cellular Ti-6Al-4V struc-
Electronic and Optical Materials, (2017), pp. 55–77 https://doi.org/10.1016/ tures with interconnected macro-porosity for bone implants fabricated by selective
B978-0-08-100433-3.00002-6. electron beam melting, Acta Biomater. 4 (2008) 1536–1544 https://doi.org/10.
[70] A. El-Hajje, E.C. Kolos, J.K. Wang, S. Maleksaeedi, Z. He, F.E. Wiria, C. Choong, 1016/j.actbio.2008.03.013.
A.J. Ruys, Physical and mechanical characterisation of 3D-printed porous titanium [95] O.L.A. Harrysson, D.R. Cormier, Direct Fabrication of Custom Orthopedic Implants
for biomedical applications material science, Mater. Med. 25 (2014) 2471–2480 Using Electron Beam Melting Technology, Advanced Manufacturing Technology
https://doi.org/10.1007/s10856-014-5277-2. for Medical Applications: Reverse Engineering, Software Conversion and Rapid
[71] T. Laoui, E. Santos, K. Osakada, M. Shiomi, M. Morita, S.K. Shaik, N.K. Tolochko, Prototyping, 2006, pp. 191–206 https://doi.org/10.1002/0470033983.ch9.
F. Abe, Properties of titanium implant models made by laser processing, Proc. IME [96] X. Gong, T. Anderson, K. Chou, Review on powder-based electron beam additive
C J. Mech. Eng. Sci. 6 (2006) 857–863 https://doi.org/10.1243/ manufacturing technology, Manuf. Rev. 1 (2014) 507–515 https://doi.org/10.
09544062JMES133. 1051/mfreview/2014001.
[72] T.D. Ngo, A. Kashani, G. Imbalzano, K.T.Q. Nguyen, D. Hui, Additive manu- [97] W.Q. Toh, P. Wang, X. Tan, M.L.S. Nai, E. Liu, S.B. Tor, Microstructure and wear
facturing (3D printing): a review of materials, methods, applications and chal- properties of electron beam melted Ti-6Al-4V Parts: a comparison study against as-
lenges, Compos. B Eng. 143 (2018) 172–196 https://doi.org/10.1016/j. cast form, Metals 6 (2016) 284 https://doi.org/10.3390/met6110284.
compositesb.2018.02.012. [98] S. Biamino, A. Penna, U. Ackelid, S. Sabbadini, O. Tassa, P. Fino, M. Pavese,
[73] F.P.W. Melchels, J. Feijen, D.W. Grijpma, A review on stereolithography and its P. Gennaro, C. Badini, Electron beam melting of Ti-48Al-2Cr-2Nb alloy: micro-
applications in biomedical engineering, Biomaterials 31 (2010) 6121–6130 structure and mechanical properties investigation, Intermetallics 19 (2010)
https://doi.org/10.1016/j.biomaterials.2010.04.050. 776–781 https://doi.org/10.1016/j.intermet.2010.11.017.
[74] J.S. Chohan, R. Singh, K.S. Boparai, R. Penna, F. Fraternali, Dimensional accuracy [99] W.P. Syam, H.A. Al-Shehri, A.M. Al-Ahmari, K.A. Al-Wazzan, M.A. Mannan,
analysis of coupled fused deposition modeling and vapour smoothing operations Preliminary fabrication of thin-wall structure of Ti6Al4V for dental restoration by
for biomedical applications, Compos. B Eng. 117 (2017) 138–149 https://doi.org/ electron beam melting, Rapid Prototype 18 (2012) 230–240 https://doi.org/10.
10.1016/j.compositesb.2017.02.045. 1108/13552541211218180.
[75] A.K. Sood, R.K. Ohdar, S.S. Mahapatra, Parametric appraisal of mechanical [100] F. Luca, M. Emanuele, R. Pierfrancesco, M. Alberto, Microstructure and mechan-
property of fused deposition modelling processed parts, Mater. Des. 31 (2010) ical properties of Ti-6Al-4V produced by electron beam melting of pre-alloyed
287–295 https://doi.org/10.1016/j.matdes.2009.06.016. powders, Rapid Prototyp. 15 (2009) 171–178 https://doi.org/10.1108/
[76] O.A. Mohamed, S.H. Masood, J.L. Bhowmik, Optimization of fused deposition 13552540910960262.
modeling process parameters: a review of current research and future prospects, [101] L.E. Murr, S.M. Gaytan, F. Medina, E. Martinez, J.L. Martinez, D.H. Hernandez,
Adv. Manufact. 3 (2015) 42–53 https://doi.org/10.1007/s40436-014-0097-7. et al., Characterization of Ti-6Al-4V open cellular foams fabricated by additive
[77] S.A.M. Tofail, E.P. Koumoulos, A. Bandyopadhyay, S. Bose, L. O'Donoghue, manufacturing using electron beam melting, Mater. Sci. Eng. 527 (2010)
C. Charitidis, Additive manufacturing: scientific and technological challenges, 1861–1868 https://doi.org/10.1016/j.msea.2009.11.015.
market uptake and opportunities, Mater. Today 21 (2018) 22–37 https://doi.org/ [102] W.S.W. Harun, M.S.I.N. Kamariah, N. Muhamad, S.A.C. Ghani, F. Ahmad,
10.1016/j.mattod.2017.07.001. Z. Mohamed, A review of powder additive manufacturing processes for metallic
[78] A. Bandyopadhyay, S. Bose, S. Das, 3D printing of biomaterials, MRS Bull. 40 biomaterials, Powder Technol. 327 (2018) 128–151 https://doi.org/10.1016/j.
(2015) 108–115. powtec.2017.12.058.
[79] H. Lee, C.H.J. Lim, M.J. Low, N. Tham, V.M. Murukeshan, Y.J. Kim, Lasers in [103] S. Bose, D. Ke, H. Sahasrabudhe, A. Bandyopadhyay, Additive manufacturing of
additive manufacturing: a review, Int. J. Precis. Eng. Manufact.-Green Technol. 4 biomaterials, Prog. Mater. Sci. 93 (2018) 45–111 https://doi.org/10.1016/j.
(2017) 307–322 https://doi.org/10.1007/s40684-017-0037-7. pmatsci.2017.08.003.
[80] J.Z. Manapat, Q.Y. Chen, P.R. Ye, R.C. Advincula, 3D Printing of polymer nano- [104] B.V. Krishna, S. Bose, A. Bandyopadhyay, Low stiffness porous Ti structures for
composites via stereolithography, Macromol. Mater. Eng. (2017) 1–13 https://doi. load-bearing implants, Acta Biomater. 3 (2007) 997–1006 https://doi.org/10.
org/10.1002/mame.201600553. 1016/j.actbio.2007.03.008.
[81] I. Gibson, D. Rosen, B. Stucker, Directed Energy Deposition Processes, Additive [105] Y.C. Wu, C.N. Kuo, M.Y. Shie, Y.L. Su, L.J. Wei, S.Y. Chen, J.C. Huang, Structural
Manufacturing Technologies: 3D Printing, Rapid Prototyping, and Direct Digital design and mechanical response of gradient porous Ti-6Al-4V fabricated by elec-
Manufacturing, Springer New York, New York, NY, 2015, pp. 245–268. tron beam additive manufacturing, Mater. Des. 158 (2018) 256–265 https://doi.
[82] W.Y. Yeong, N. Sudarmadji, H.Y. Yu, C.K. Chua, K.F. Leong, S.S. Venkatraman, org/10.1016/j.matdes.2018.08.027.
Y.C.F. Boey, L.P. Tan, Porous polycaprolactone scaffold for cardiac tissue en- [106] A. Ataee, Y. Li, M. Brandt, C. Wen, Ultrahigh-strength titanium gyroid scaffolds
gineering fabricated by selective laser sintering, Acta Biomater. 6 (2010) manufactured by selective laser melting (SLM) for bone implant applications, Acta
2028–2034 https://doi.org/10.1016/j.actbio.2009.12.033. Mater. 158 (2018) 354–368 https://doi.org/10.1016/j.actamat.2018.08.005.
[83] J.P. Kruth, B. Vandenbroucke, V.J. Vaerenbergh, P. Mercelis, Benchmarking of [107] M.A. Surmeneva, R.A. Surmenev, E.A. Chudinova, A. Koptioug, M.S. Tkachev,
Different SLS/SLM Processes as Rapid Manufacturing Techniques, Division PMA, S.N. Gorodzha, L. Rännar, Fabrication of multiple-layered gradient cellular metal
Department of Mechanical Engineering, Katholieke Universiteit Leuven, Belgium, scaffold via electron beam melting for segmental bone reconstruction, Mater. Des.
2005. 133 (2017) 195–204 https://doi.org/10.1016/j.matdes.2017.07.059.
[84] Y. Liu, Y. Yang, S. Mai, D. Wang, C. Song, Investigation into spatter behavior [108] X.Z. Xin, N. Xiang, J. Chen, B. Wei, In vitro biocompatibility of Co–Cr alloy fab-
during selective laser melting of AISI 316L stainless steel powder, Mater. Des. 87 ricated by selective laser melting or traditional casting techniques, Mater. Lett. 88
(2015) 797–806 https://doi.org/10.1016/j.matdes.2015.08.086. (2012) 101–103 https://doi.org/10.1016/j.matlet.2012.08.032.

69
L. Yuan et al. Bioactive Materials 4 (2019) 56–70

[109] A.G. Demir, B. Previtali, Additive manufacturing of cardiovascular CoCr stents by [123] C. Yan, L. Hao, A. Hussein, P. Young, Ti–6Al–4V triply periodic minimal surface
selective laser melting, Mater. Des. 119 (2017) 338–350 https://doi.org/10.1016/ structures for bone implants fabricated via selective laser melting, Mech. Behav.
j.matdes.2017.01.091. Biomed. Mater. 51 (2015) 61–73 https://doi.org/10.1016/j.jmbbm.2015.06.024.
[110] E.A. Lord, A.L. Mackay, S. Ranganatan, New Geometry for New Materials, [124] V.I. Sikavitsas, J.S. Temenoff, A.G. Mikos, Biomaterials and bone mechan-
Cambridge University Press, 2006. otransduction, Biomaterials 22 (2001) 2581–2593 https://doi.org/10.1016/
[111] M.M. Sychov, L.A. Lebedev, S.V. Dyachenko, L.A. Nefedova, Mechanical proper- S0142-9612(01)00002-3.
ties of energy-absorbing structures with triply periodic minimal surface topology, [125] O. Al-Ketan, R.K.A. Al-Rub, R. Rowshan, Mechanical properties of a new type of
Acta Astronaut. (2017) 1–4 https://doi.org/10.1016/j.actaastro.2017.12.034. architected interpenetrating phase composite materials, Adv. Mater. Technol. 2
[112] D.J. Yoo, Porous scaffold design using the distance field and triply periodic (2017), https://doi.org/10.1002/admt.201600235.
minimal surface models, Biomaterials 32 (2011) 7741–7754 https://doi.org/10. [126] S. Rajagopalan, R.A. Robb, Schwarz meets Schwann: design and fabrication of
1016/j.biomaterials.2011.07.019. biomorphic and durataxic tissue engineering scaffolds, Med. Image Anal. 10
[113] F. Conan, 3D-printed porous bed structures, Curr. Opin. Chem. Eng. 18 (2017) (2006) 693–712 https://doi.org/10.1016/j.media.2006.06.001.
10–15 https://doi.org/10.1016/j.coche.2017.07.003. [127] D.J. Yoo, K.H. Kim, An advanced multi-morphology porous scaffold design method
[114] D. Li, W. Liao, N. Dai, G. Dong, Y. Tang, Y.M. Xie, Optimal design and modeling of using volumetric distance field and beta growth function, Int. J. Precis. Eng.
gyroid-based functionally graded cellular structures for additive manufacturing, Manuf. 16 (2015) 2021–2032 https://doi.org/10.1007/s12541-015-0263-2.
Comput. Aided Des. 104 (2018) 87–99 https://doi.org/10.1016/j.cad.2018.06. [128] C.Z. Yan, L. Hao, A. Hussein, D. Raymont, Evaluations of cellular lattice structures
003. manufactured using selective laser melting, Int. J. Mach. Tool Manufact. 62 (2012)
[115] F.P.W. Melchels, K. Bertoldi, R. Gabbrielli, A.H. Velders, J. Feijen, D.W. Grijpma, 32–38 https://doi.org/10.1016/j.ijmachtools.2012.06.002.
Mathematically Defined Tissue Engineering Scaffold Architectures Prepared by [129] A.H. Scheon, Reflections concerning triply-periodic minimal surfaces, Interf. Focus
Stereolithography Biomaterials, (2010), pp. 6909–6916 31 https://doi.org/10. 2 (2012) 658–668 https://doi.org/10.1098/rsfs.2012.0023.
1016/j.biomaterials.2010.05.068. [130] A.H. Schoen, Infinite Periodic Minimal Surfaces without Self-intersection, National
[116] K3dsurf. http://k3dsurf.sourceforge.net/, (Accessed 1 May 2018). Aeronautics and Space Administration (NASA) Tech, Washington, DC, 1970 Note
[117] M. Speirs, B. Van Hooreweder, J. Van Humbeeck, J.P. Kruth, Fatigue behaviour of no. D-5541.
NiTi shape memory alloy scaffolds produced by SLM, a unit cell design compar- [131] S.C. Kapfer, S.T. Hyde, K. Mecke, C.H. Arns, G.E. Schroder-Turk, Minimal surface
ison, Journal of the Mechanical Behavior of Biomedical Materials 70 (2016) 53–59 scaffold designs for tissue engineering, Biomaterials 32 (2011) 6875–6882 https://
https://doi.org/10.1016/j.jmbbm.2017.01.016. doi.org/10.1016/j.biomaterials.2011.06.012.
[118] F.S.L. Bobbert, K. Lietaert, A.A. Eftekhari, B. Pouran, S.M. Ahmadi, H. Weinans, [132] I. Maskery, N.T. Aboulkhair, A. Aremu, C. Tuck, I. Ashcroft, Compressive failure
A.A. Zadpoor, Additively manufactured metallic porous biomaterials based on modes and energy absorption in additively manufactured double gyroid lattices,
minimal surfaces: a unique combination of topological, mechanical, and mass Add. Manufact. 16 (2017) 24–29 https://doi.org/10.1016/j.addma.2017.04.003.
transport properties, Acta Biomater. 53 (2017) 572–584 https://doi.org/10.1016/ [133] O. Al-Ketan, R. Rowshan, R.K. Abu Al-Rub, Topology-mechanical property rela-
j.actbio.2017.02.024. tion- ship of 3D printed strut, skeletal, and sheet based periodic metallic cellular
[119] T. Femmer, J.C.K. Alexander, M. Wessling, Estimation of the structure dependent materials, Add. Manufact. 19 (2018) 167–183 https://doi.org/10.1016/j.addma.
performance of 3-D rapid prototyped membranes, Chem. Eng. J. 273 (2015) 2017.12.006.
438–445 https://doi.org/10.1016/j.cej.2015.03.029. [134] K. Monkova, P. Monka, I. Zetkova, P. Hanzl, D. Mandulak, Three approaches to the
[120] B.D. Nguyen, Y.C. Jeong, K. Kang, Design of the P-surfaced shellular, an ultra-low gyroid structure modelling as a base of lightweight component produced by ad-
density material with micro-architecture, Comput. Mater. Sci. 139 (2017) ditive technology, 2017 2nd International Conference on Computational
162–178 https://doi.org/10.1016/j.commatsci.2017.07.025. Modeling, Simulation and Applied Mathematics.
[121] H.A. Almeida, P.J. Bartolo, Design of tissue engineering scaffolds based on hy- [135] A. Yánez, A. Herrera, O. Martel, D. Monopoli, H. Afonso, Compressive behaviour
perbolic surfaces: structural numerical evaluation, Med. Eng. Phys. 36 (2014) of gyroid lattice structures for human cancellous bone implant applications, Mater.
1033–1040 https://doi.org/10.1016/j.medengphy.2014.05.006. Sci. Eng. C 68 (2016) 445–448 https://doi.org/10.1016/j.msec.2016.06.016.
[122] S. Gómez, M.D. Vlad, J. López, E. Fernández, Design and properties of 3D scaffolds [136] V.J. Challis, X. Xu, L.C. Zhang, A.P. Roberts, J.F. Grotowski, T.B. Sercombe, High
for bone tissue engineering, Acta Biomater. 42 (2016) 341–350 https://doi.org/ specific strength and stiffness structures produced using selective laser melting,
10.1016/j.actbio.2016.06.032. Mater. Des. 63 (2014) 783–788 https://doi.org/10.1016/j.matdes.2014.05.064.

70