Materials Science and Engineering C 53 (2015) 50–59

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Porous niobium coatings fabricated with selective laser melting on titanium substrates: Preparation, characterization, and cell behavior Sheng Zhang a,b,1, Xian Cheng c,1, Yao Yao c, Yehui Wei c, Changjun Han b, Yusheng Shi b, Qingsong Wei b,⁎, Zhen Zhang b,c,⁎⁎ a b c

Science and Technology on Power Beam Processes Laboratory, Beijing Aeronautical Manufacturing Technology Research Institute (BAMTRI), Beijing 100024, China State Key Lab of Materials Processing and Die & Mould Technology, Huazhong University of Science and Technology, Wuhan 430074, China Department of Stomatology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China

a r t i c l e

i n f o

Available online 14 April 2015 Keywords: Niobium Titanium Selective laser melting Coatings Biocompatibility

a b s t r a c t Nb, an expensive and refractory element with good wear resistance and biocompatibility, is gaining more attention as a new metallic biomaterial. However, the high price of the raw material, as well as the high manufacturing costs because of Nb's strong oxygen affinity and high melting point have limited the widespread use of Nb and its compounds. To overcome these disadvantages, porous Nb coatings of various thicknesses were fabricated on Ti substrate via selective laser melting (SLM), which is a 3D printing technique that uses computer-controlled high-power laser to melt the metal. The morphology and microstructure of the porous Nb coatings, which had pores ranging from 15 to 50 μm in size, were characterized with scanning electron microscopy (SEM). The average hardness of the coating, which was measured with the linear intercept method, was 392 ± 37 HV. In vitro tests of the porous Nb coating which was monitored with SEM, immunofluorescence, and CCK-8 counts of cells, exhibited excellent cell morphology, attachment, and growth. The simulated body fluid test also proved the bioactivity of the Nb coating. Therefore, these new porous Nb coatings could potentially be used for enhanced early biological fixation to bone tissue. In addition, this study has shown that SLM technique could be used to fabricate coatings with individually tailored shapes and/or porosities from group IVB and VB biomedical metals and their alloys on stainless steel, Co–Cr, and other traditional biomedical materials without wasting raw materials. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Recently, new metallic biomaterials based on the expensive and refractory elements from IVB and VB groups such as Ta, Zr, and Nb, which have good wear resistance, are gaining more attention [1]. In vitro apatite formation test and in vivo histomorphometric study have shown that these materials are bioactive and can biologically bond to bone [2,3]. However, the high price of the raw materials and relatively high manufacturing costs to produce pure Ta or Nb implants have limited their widespread use [4]. The extremely high melting point of Nb (2468 °C), as well as its high affinity for oxygen, makes it difficult to fabricate Nb implants via conventional processing routes. Lately, the metal additive manufacturing (MAM) process, a 3D printing technique, offers a nearly unlimited flexibility by using high-power laser beam, and provides special opportunities for orthopedic and bone implants [5].

⁎ Corresponding author. ⁎⁎ Correspondence to: Z. Zhang, State Key Lab of Materials Processing and Die & Mould Technology, Huazhong University of Science and Technology, Wuhan 430074, China. E-mail addresses: [email protected] (Q. Wei), [email protected] (Z. Zhang). 1 Contributed equally to this work.

http://dx.doi.org/10.1016/j.msec.2015.04.005 0928-4931/© 2015 Elsevier B.V. All rights reserved.

Compared with other MAM techniques, selective laser melting (SLM) allows layers and coatings of other refractory metal materials to be fabricated on the surface of metal implants [6]. A schematic of a SLM set-up is shown in Fig. 1. It is more accurate than other MAM techniques because it uses fiber laser, which possesses smaller spot sizes (b 100 μm) and highly concentrated output energies compared with other lasers [7]. By combining the high-energy fiber laser with a computercontrolled laser beam system [8,9], this gives the ability to control the melting of Nb, and thus, the ability to create implants with tailored shapes and/or porosities. Further details about the SLM procedure and a description of the parameters have been reported in a previous study [10]. Recently, SLM has been used to fabricate common, metal-based medical materials, which contain Ti and its alloys, Co–Cr alloys, stainless steel, and others [11], that have been widely used for orthopedic and dental implants [12,13]. In this paper, we report the first example of an irregular, porous, Nb coating fabricated directly with SLM on pure Ti. We investigated the fabrication of porous Nb coatings on pure Ti, and performed in vitro studies to assess the cell attachment, morphology, and proliferation of the samples. These new porous Nb coatings could potentially be used for enhanced early biological fixation to bone tissue. In addition, this study has shown that SLM can be used to fabricate coatings with

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

51

Fig. 1. Schematic of the selective laser melting (SLM) set-up.

individually tailored shapes and/or porosities from group IVB and VB biomedical metals and their alloys on stainless steel, Co–Cr, and other traditional biomedical materials, without wasting raw material. 2. Materials and methods 2.1. Materials The materials used were Nb powder (99.5% purity, particle size = 15–125 μm, Xing Rongyuan Technology Co. Ltd., Beijing, China), and commercially available pure Ti (grade 2, Baoji Ti Industry Co., Ltd., Baoji, China). MC3T3-E1 osteoblast-like cells (ATCC catalog CRL-2593) were obtained from the American Type Culture Collection. All the other chemical reagents were local products of analytical grade. 2.2. Fabrication of the porous Nb coating A porous Nb coating (thickness = 2.0 mm, diameter = 8 mm) was deposited on a Ti substrate with a SLM machine (HRPM-II, Rapid Manufacturing Center, Huazhong University of Science and Technology, China). The SLM processing was performed with a 200 W fiber laser (YLP-HP, IPG Photonics Corporation, Germany) in a high-purity argon atmosphere to limit the oxidation of Nb and Ti in the fabrication chamber. After process optimization, a laser power of 160 W, scanning speed of 200 mm/s, hatching spacing of 0.07 mm, and layer thickness of 0.04 mm were applied to melt Nb onto the Ti substrate. The optimization of the processing parameters ensured that the residual stress was not large enough to cause warping of the thin Nb coating. In addition, special scanning strategies were adopted to fabricate the porous Nb coatings. Before the fabricating process, 3D CAD data was prepared and then the powder was sliced into layers with a defined thickness in

the vertical direction. As shown in Fig. 2, the scan lines of each subsequent layer are perpendicular to the previous layer. Therefore, the SLM process involves a series of repetitive procedures that form a layer of powder and then transfers specific geometric information onto the material by melting the powder with a laser beam. Under these processing conditions, the porous Nb coating was found to be visually sound, with a strong adherence to the Ti substrate. 2.3. Characterization of the porous Nb coating The cross-sectional microstructure of the samples was observed with optical microscopy and field emission scanning electron microscopy (FE-SEM, JEOL-JSM7600F, Japan). The chemical composition of the Nb powder was measured with dispersive X-ray spectroscopy (EDS, X-Max 50, INCA, England). To reveal microstructural features, the chemically polished samples were then etched for 60 s with a solution containing 20 ml of HNO3, 20 ml of HF, and 60 ml of H2SO4. To track any diffusion that may occur at the interface between the porous Nb coating and Ti substrate, a series of micro-hardness indentations were applied to the deposited Nb from one end to the other, with adjacent indents of 0.05 mm spacing, using a Vickers micro-hardness tester (HXS-1000AK, Shanghai, China) with a load of 100 g for 15 s. 2.4. In vitro tests MC3T3-E1 osteoblast-like cells were cultured in an alpha minimum essential medium (α-MEM, Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Grand Island, NY, USA) at 37 °C in a humidified, 5% CO2 atmosphere. When the cells reached 80–90% confluence, they were trypsinized and suspended in the culture medium. Having been steam sterilized and placed in 24-well tissue culture plates

Fig. 2. A schematic of special scanning strategies used to fabricate the porous Nb coating on pure Ti.

52

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

under aseptic conditions, the Nb coatings and Ti substrates were seeded at a density of 2000 cells·cm−2. To observe the F-actins of the osteoblasts on the porous Nb, three samples were fixed with a 4% paraformaldehyde solution after 3 and 7 days of culture, then washed with phosphate buffered saline (PBS), and finally permeabilized with a 0.1% Triton X-100 solution for 5 min. The non-specific binding sites were blocked by incubating the coating for 30 min in a PBS solution that contained 1% bovine serum albumin. The F-actins were stained with rhodamine phalloidin (R-415 kit, Molecular Probes, Invitrogen, USA) for 20 min at room temperature, and the nuclei were stained with 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI, 1:1000 dilutions in PBS, Invitrogen, Basel, Switzerland) for 15 min. The samples were rinsed again with PBS and then stored in a 9:1 mixture of glycerol/PBS. Immunofluorescence images were obtained with a Nikon TE-2000 inverted microscope. Fluorescence images of the two stains were taken by double exposure, with the red exposure followed by the blue exposure. To observe the cell morphology, three samples were examined with SEM. At predetermined time intervals (3 and 7 days), the samples were gently washed 3 times with PBS and then fixed with a 2.5% glutaraldehyde solution for 24 h at 4 °C. Having been thoroughly washed with PBS, the samples were dehydrated with a graded ethanol series, sputter-coated with gold in a vacuum, and then examined with SEM after critical point drying. Cell proliferation was quantitatively analyzed with cell counting kit-8 (CCK-8, Dojindo Laboratories, Japan). At specific time intervals (1, 3, and 7 days), some samples were gently washed with PBS, and then 0.5 ml of an α-MEM solution containing 10% CCK-8 was added to each well. The samples were incubated at 37 °C for 3 h. The absorbance of the supernatant was then measured at 450 nm with an ELX808 Ultra Microplate Reader (Bio-Tek Instruments, Inc., USA), and the determination of the optical density values was performed at minimum in triplicate, which reflected the viable cell population in each well. A modified simulated body fluid (SBF) recipe proposed by Oyane et al. was used for the apatite deposition tests in the present study

[20]. The SBF was prepared by dissolving the following chemicals in the sequence given: NaCl (5.403 g), NaHCO3 (0.504 g), Na2CO3 (0.426 g), KCl (0.225 g), K2HPO4·3H2O (0.230 g), MgCl2·6H2O (0.311 g), CaCl2 (0.293 g), and Na2SO4 (0.072 g). The mixture was then buffered to a pH of 7.40 with 4-(2-hydroxyethyl)-piperazine-1ethanesulfonic acid (HEPES) and a 1 M solution of NaOH at 37 °C. The Nb coating and Ti substrates were placed in a polyethylene (PE) Petri dish with 30 ml of SBF, which was kept in an incubator at 37 °C for 2 and 4 weeks with the SBF refreshed every 2 days to keep the ion concentration stable. The precipitation particles deposited on the surfaces of the Nb coating and Ti substrate were measured with EDS. After each period, samples were removed from the SBF, rinsed with distilled water, left to dry in air, and then stored in a desiccator for further characterization.

3. Results 3.1. Morphology and composition of the Nb powder The SEM images in Fig. 3a show the characteristic morphology of the Nb powder which contains irregularly shaped particles with an average size of 75 μm. Unlike spherical particles, the irregularly shaped Nb particles may present an inherent eccentricity, and thus, a torque could be formed during the SLM process, which is beneficial for the formation of porous structures. As can be seen from the surface microstructure of the raw Nb powder (Fig. 3b), there are some loose structures on the surface of the particle. The EDX area scan (Fig. 3c) reveals that the surface of the Nb powder mainly consists of Nb and O with an atomic ratio of 2.156, which implies the presence of Nb2O5 and NbO2. In addition, the EDX point analysis (Fig. 3d) shows that Nb and O are the major elements present with an Nb/O atomic ratio of approximately 2.27, which suggests that the white structures on the surface of the particles contain more of the Nb2O5 phase. These results demonstrate that the raw Nb

Fig. 3. SEM images showing the (a) characteristic morphology of the Nb powder and (b) its surface morphology under high-magnification. (c) The chemical composition of the Nb powder in Fig. 3b as determined with an EDX area scan. (d) EDX point analysis of the white structure marked in Fig. 3b.

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

53

powder does not contain any potential impurities that could have adverse effects on biological activity. 3.2. Characteristic morphology of the porous Nb coating The morphology of the SLM-made, porous, Nb coating on the Ti substrate was characterized with optical microscopy and SEM (Fig. 4). The porosity of the coating is dependent on the irregularly shaped particles of the Nb powder and the processing parameters. The images in Fig. 4 show that the Nb powder has melt completely without any signs of the balling phenomenon that may occur during the laser melting process. The uneven contoured surface shown in Fig. 4b increases the surface roughness of the sample. It can be seen from Fig. 4c that the circular and cubic micro-pores are distributed randomly throughout the Nb coating, which may be a result of the irregularly shaped Nb particles. As can be seen from the exterior face of a vertical slice of the sample (Fig. 4d), the porous Nb coating exhibits no interlayer differentiation, which indicates good metallurgical bonding between the individual layers because there are no cracks or fusion defects. Therefore, the laser energy input used was sufficient for the complete melting of the Nb powder and strong bonding between the Nb layer and Ti substrate during the SLM fabrication process. 3.3. Microstructure and hardness of the Nb coating The average grain size of the porous Nb coating was 70 ± 20 μm (Fig. 5), which was determined with the linear intercept method. The grains of the Nb coating are irregularly shaped, and could be easily distinguished from the microstructure of the surrounding Ti. However, the Nb and Ti grains appeared to be bonded together without any

Fig. 5. The microstructure of the Nb coating formed by SLM.

obvious cracks, indicating good bonding. The micro-hardness of the sample was tested at various distances from Ti substrate. The microhardness test conformed to the standards required for measuring Vickers hardness. The average hardness of the SLM-fabricated, porous, Nb coating is 392 ± 37 HV (Fig. 6), which is higher than that of Ti substrate (189 ± 4 HV). As can be seen from Fig. 6, the microhardness increases as the distance from the location tested to the Ti–Nb interface increases.

Fig. 4. Optical microscopy and SEM images showing the morphology of the porous Nb coating on a Ti substrate: (a) Ti substrate with the Nb coating, (b) the surface of the porous Nb coating, (c) a cross section of the porous Nb coating, and (d) a vertical section of the porous Nb coating and Ti substrate.

54

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

from adjacent cells can be easily seen, showing the communication among the cells. After 7 days of culture (Fig. 7b and d), the number of the cell nuclei showing a bright blue fluorescence has increased, indicating that the number of cells has increased as the period of culture on both surfaces increased. Moreover, the cells on both surfaces also have become better spread and have more filopodia and lamellipodia after 7 days of culture compared with that of 3 days of culture.

3.5. Cell morphology and growth

Fig. 6. Micro-hardness profile of the SLM-processed, porous, and Nb coating on a Ti substrate.

3.4. Cell attachment and cytoskeleton The influence of Nb and Ti on the cytoskeleton and cell attachment was investigated over a certain period of cell culture with the samples checked after 3 and 7 days. As can be seen from Fig. 7a and c, after 3 days of culture, the osteoblast-like cells cultured on both the Nb coating and Ti substrate have pronounced and elongated filopodia projecting from the cell edges. In addition, the connection of protrusions

SEM images showing the morphology and growth of the cells on the Nb coating and Ti substrate after being cultured for 7 and 11 days are shown in Figs. 8 and 9. On the Nb coating, flattened and well-spread cells with plenty of micro-extensions can be seen after 7 days of culture (Fig. 8a), and the cells have adhered to each other through abundant and well-developed connections of cellular protrusions (Fig. 8b). In addition, the cells have begun to migrate into the pores of the Nb coating after 7 days of culture (Fig. 8a). By contrast, the cells on the Ti surface have become more rounded in shape and spread with less micro-extensions (Fig. 8d). Moreover, the cell growth is less noticeable and there is an inadequate establishment of cell spreading (Fig. 8c). The SEM images of the cells cultured for 11 days are presented in Fig. 9, from which differences between the cell morphology and growth of the samples can be seen. On the Nb coating, the number of cells has increased (Fig. 9a), and the entire surface of the sample is covered with flat and well-spread cells with numerous micro-extensions and well-developed connections of cellular protrusions (Fig. 9b). The cells have grown in a confluent multilayer on the Nb coating, which implies

Fig. 7. Fluorescence images of the cytoskeleton and cell attachment of osteoblast-like cells after being cultured on the (a–b) porous Nb coating and (c–d) Ti substrate for 3 and 7 days. DAPI has stained the nuclei blue and rhodamine phalloidin has stained the actin filaments red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

Fig. 8. SEM images showing the cell morphology and growth of osteoblast-like cells after being cultured on the (a–b) porous Nb coating and (c–d) Ti substrate for 7 days.

Fig. 9. SEM images showing the cell morphology and growth of osteoblast-like cells after being cultured on the (a–b) porous Nb coating and (c–d) Ti substrate for 11 days.

55

56

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

enhanced extracellular matrix (ECM) formation (Fig. 9a), and the cellular surfaces on the Nb coating are rougher, indicating some level of ECM mineralization (Fig. 9b). In addition, where the cells on the Nb coating have spread and grown more, they have bridged the pores (Fig. 9a). However, after 11 days of culture, less cell spreading can be seen on the Ti substrate (Fig. 9d), and the cells even have not formed a complete confluent layer (Fig. 9c). 3.6. Cell proliferation The number of the living cells, a determinant for the level of cell proliferation on the laser-processed Nb coating and Ti substrate, was determined by a CCK-8 assay, with the results shown in Fig. 10. After 3 and 7 days of culture, there is no significant difference (p N 0.05) between the CCK-8 cell counts of the Nb coating and the Ti substrate. However, after 11 days of culture, CCK-8 cell count of the Nb coating is significantly higher than that of the Ti substrate (p b 0.05). 3.7. Bioactivity of the Nb coating Fig. 11b and e shows SEM images of a layer of precipitation spheroids deposited on the surfaces of the Nb coating and Ti substrate, respectively, after being soaked in SBF for 2 weeks. The EDX area scan results (Fig. 12) imply that the precipitation particles are mainly composed of oxygen, calcium, and phosphorus. As the soaking time increases to 4 weeks, the layers have become thicker and denser (Fig. 11c and f). Moreover, it can be seen in the SEM images that the Nb coating is completely covered with multiple layers of precipitation particles (Fig. 11c), while the Ti substrate is covered with a single layer (Fig. 11f). 4. Discussion 4.1. Production and characterization The laser absorptivity of a material directly controls the amount of energy transferred from the laser to the sample. The laser absorption coefficient of a material is directly proportional to its electrical resistance and inversely proportional to the wavelength of the laser [4]. The relatively low electrical resistivity of Nb (12.5 × 10− 8 Ω·m) compared with that of Ti (40 × 10−8 Ω·m) decreases the laser absorption coefficient of Nb. This results in the incomplete melting of the Nb powder, which is especially true for the coarser particles in the Nb

Fig. 10. Results of the CCK-8 assay that quantitatively measured the number of living osteoblast-like cells after being cultured on the porous Nb coating (red) and Ti substrate (blue) for 3, 7, and 11 days. The error bars represent the mean ± standard deviation for n = 4 (*p b 0.05). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

powder used for this study. The high melting point of Nb (2468 °C) makes it difficult to melt the powder completely during the extremely short time of the SLM process. As shown in Fig. 3, the irregular pores in the Nb coating are obvious. In general, the pores on the surface can improve the wettability and surface energy of a material, which are beneficial for enhanced early biological fixation [1]. Usually, the solidified structures in SLM-processed materials are finely sized because of the fast cooling rate of this method. However, in the present study, the average grain size of the porous Nb coating was found to be 70 ± 20 μm. This is due to the high laser energy input used to melt the Nb coating onto Ti substrate, which results in a slower cooling rate because of the shallow thermal gradient during deposition. These coarse grains can reduce the hardness of Nb and enhance its adaptability to human bone. The variation of grain size across the coating leads to the large variation in the hardness of the Nb coating (Fig. 6). The hardness profile across the coating shows that the hardness of Nb gradually increases from 250 HV at the interface to 400 HV in the region just below the outer surface. The high hardness of Ti (263 HV) just below the coating interface is because of the solid solution strengthening, which is caused by the dissolution of Nb in Ti. The smooth nature of the hardness profile, without any abrupt increases in hardness on either side of the interface, implies the absence of undesirable phases that could form due to the reactions between Nb and Ti. These results clearly demonstrate that the SLM process can produce porous Nb coatings on pure Ti, and the technique can be extended to fabricate net-shaped Nb structures if necessary. 4.2. Analysis of the in vitro cell behavior The interaction between biomaterials and bone tissue is a key factor for bone substitution materials. To provide qualitative information on the cytoskeleton and cell attachment on the Nb coatings and Ti substrates, fluorescence images were taken after being cultured for 3 and 7 days to monitor the formation of the actin cytoskeleton, which provides a structural framework and participates in cell attachment [14]. After 3 days of culture, cells on both Nb and Ti surfaces had extended lamellipodia and filopodia, implying that the Nb coating possesses an attachment between the cells and material surface that is as strong as the Ti substrate. In previous studies, filopodia have been found to first sense the surface topography and then form attachments which allow the cells to become strongly attached to the surface. The connection of cellular protrusions with both Nb and Ti surfaces indicated active communication between adjacent cells. Davenport et al. [15] found that the filopodia and connections of cellular protrusions provide the routes for transporting nutrients and exchanging information, which plays an important role in cell communication. After 7 days of culture, on both Nb and Ti surfaces, there was an increase in the number of cells. In addition, the cells had spread better and had more well-developed lamellipodia, filopodia, and connections of cellular protrusions compared with the sample after 3 days. This indicates that the Nb coatings can promote cell attachment and initial proliferation to the same degree as the Ti substrate. For osseointegration, the cell attachment and subsequent cellular responses to the biomaterials are the critical parameters [16]. To provide qualitative information on the cell morphology and growth on the porous Nb coating and Ti substrate, SEM images were taken after being cultured for 7 and 11 days. After 7 days of culture, the cells on the Nb coating had a three-dimensional flat morphology with numerous micro-extensions and well-developed connections of cellular protrusions, which suggests significantly better cell spreading and growth. On the other hand, the cells on the Ti surface appeared round and exhibited few micro-extensions and connections of cellular protrusions, which indicates that they had failed to spread and grow well on the Ti surface. Qin et al. found that cells are well-spread and strongly attached when their morphology is no longer round and the formation of protrusions and cell flattening are evident [17]. Krishna

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

57

Fig. 11. SEM images showing a layer of precipitation particles deposited on the surfaces of the (b–c) Nb coating and (e) (f) Ti substrate after being soaked in SBF for 2 and 4 weeks. The control samples of (a) Nb and (d) Ti were not soaked in SBF.

et al. found that more micro-extensions and well-developed connections of cellular protrusions are evidence of favorable surface characteristics for cell communication and growth, especially in the early periods of cell culture [1]. Although, the number of cells on both Nb and Ti surfaces increased with the increasing culture time, the Ti surface still showed less cell spread and did not form a complete confluent layer, even after 11 days of culture. On the other hand, the cells on the Nb coating became flatter and well-spread, and formed dense and confluent cellular layers, implying enhanced ECM formation. In addition, rougher cellular surfaces were found on the Nb coating indicating some level of ECM mineralization, which could be interpreted as an early stage of osteoblast differentiation. These differences between the cell morphology and growth on the Nb and Ti surfaces may be explained by the different surface characteristics of these materials, such as wettability, surface energy, pore size, and pore shape. The rough surface topography of a porous structure can enhance the wettability [18], and pure Nb has been reported to have a higher surface energy (2.37 J/m2) than that of

pure Ti (1.24 J/m2) [19,20], indicating that porous Nb coatings have a higher surface energy and wettability than Ti substrates. Similar results showing that a surface with a higher wettability and surface energy that exhibits better cell response has been previously reported [21–23]. According to Reilly et al. [24], the cell morphology, growth, and differentiation are stimulated by the mechano-biological stimuli within a porous structure, because the pore size and shape are hypothesized to have an influence on how cells sense the environment. On Ti substrates, the cells can only feel the single stimuli of the outer surface, while in the irregularly shaped and moderately sized pores of the Nb coating, cells can touch several walls within the pores, resulting in different stimuli. Of course, this idea is still speculative and needs further investigation. However, SEM and fluorescence images were both qualitative observations of cell attachment and growth. For this reason, the CCK-8 assay, a quantitative measurement of metabolic activity, determined the cell behavior accurately. When cultured for 7 days or less, there was no significant difference (p N 0.05) between the number of living cells in the Nb and Ti samples, indicating that the cell attachment and initial

Fig. 12. Results of the EDX area scans that reveal the chemical compositions of the precipitation particles deposited on the surfaces of the (a–b) Nb coating and (c–d) Ti substrate after being soaked in SBF for 2 and 4 weeks.

58

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59

proliferation on the Nb coating are similar to that of Ti substrate. This is consistent with the qualitative fluorescence imaging observations of cell attachment and cytoskeleton. Although the CCK-8 assay results showed no statistically significant difference between the number of cells when cultured for 7 days or less, other studies have demonstrated that highly porous structures are beneficial for the bone in-growth and extensive body fluid transportation required for osseointegration [25,26]. As the culture time increased to 11 days, the number of cells on the Nb and Ti surfaces increased, and the difference between the number of cells on both surfaces was significant (p b 0.05), which shows that the porous Nb coatings can enhance the cell growth and long-term proliferation. This agrees well with the qualitative SEM results for the cell growth and morphology. The reasons for these phenomena may be threefold: (i) the porous structure favors oxygen and nutrient supply, and has more open space and a larger surface area for cells to attach to and grow on [27]; (ii) the mechano-biological stimuli ascribed to the porous structure, as discussed above in Section 4.2; and (iii) the higher surface energy and wettability inherent to Nb and the rough surface topography of the pores, as discussed above in Sections 4.1 and 4.2. Through SEM and EDX analysis, the Nb surface was found to be completely covered by multiple layers of precipitation particles composed of oxygen, calcium, and phosphorus after being soaked in SBF for 4 weeks. According to other studies [12], it is reported that the roughness change of the porous Nb surface topography formed by using a high power laser beam during the SLM process plays a crucial role. However, this mechanism needs further research. Hoshikawa et al. [28] and Pauline et al. [29] reported that the bioactivity of hydroxyapatite deposition is improved on the methylsiloxane–Ca–Nb–Ta coating on Ti derived by sol–gel, and Sr incorporated Nb2O5 coating on stainless steel by spin-coating method, respectively. However, our pure Nb coatings that show good cell response can avoid the uncertainty of cell cytotoxicity caused by the other elements and compounds that comprise these coatings. Ramírez et al. reported that NbN and Nb2O5 coatings deposited by magnetron sputtering enhanced the biological response of implants, but neither of them is better than Ti and its alloys [30], while our porous Nb coating exhibited better attachment and growth than Ti. Moreover, different from the methods for fabricating these three coatings mentioned above, the SLM process adopted to fabricate our pure Nb coating, a 3D printing technique, can not only create coatings with individually tailored shapes and/or porosities [31], but it can also produce coatings with ideal metallurgical bonding, because the metal powder is completely melted during the SLM process [32]. In addition, while the relatively high price of Nb has limited the widespread acceptance of the coatings composed of Nb and its compounds mentioned above, the SLM process, an additive manufacturing technique, can eliminate raw material waste, and thus, significantly reduce the costs of using Nb and its compounds [33]. When Ta, another refractory element from group VB as Nb, is fabricated as dense or porous coatings on Ti also by SLMs, the coatings show good biocompatibility that is similar to Nb [4]. However, the significantly lower melting temperature of Nb (2468 °C) compared with that of Ta (3017 °C) [34] makes the SLM processing of Nb coatings easier, and the continental crust has a well-defined Nb/Ta ratio of 11–12 [35], and therefore, the Nb coatings are less costly. However, extensive in vitro analysis is required to assess the cell differentiation and functionality after a longer incubation time. These studies are ongoing, and will be published in near future. 5. Conclusions Porous Nb coatings that have ideal metallurgical bonding with Ti substrate were successfully fabricated by the SLM technique. The in vitro study of cell–material interaction showed that the Nb coating has better biocompatibility for cell attachment, morphology, and growth than that of the Ti substrate. However, the direct clinical

application of the porous Nb coating potentially used to enhance early biological fixation with bone tissue is not straightforward. This study has provided experience and proposed a solution for fabricating coatings composed of other new biomedical metals and theirs alloys from groups IVB and VB on stainless steel, Co–Cr, and other traditional biomedical materials, through the SLM technique with individually tailored shapes and/or porosities and without wasting raw materials. Acknowledgments This work has been supported by the National Key Technology R&D Program of Ministry of Science and Technology of China (Grant No. 2012BAF08B03), the National Natural Science Foundation of China (No. 81300920) and the National Natural Science Foundation of China (Grant No. 51375189). References [1] V.K. Balla, S. Bodhak, S. Bose, A. Bandyopadhyay, Porous tantalum structures for bone implants: fabrication, mechanical and in vitro biological properties, Acta Biomater. 6 (2010) 3349–3359. [2] C. Johansson, T. Albrektsson, A removal torque and histomorphometric study of commercially pure niobium and titanium implants in rabbit bone, Clin. Oral Implants Res. 2 (1991) 24–29. [3] X. Wang, Y. Li, J. Lin, Y. Yamada, P. Hodgson, C. Wen, In vitro bioactivity evaluation of titanium and niobium metals with different surface morphologies, Acta Biomater. 4 (2008) 1530–1535. [4] V.K. Balla, S. Banerjee, S. Bose, A. Bandyopadhyay, Direct laser processing of a tantalum coating on titanium for bone replacement structures, Acta Biomater. 6 (2010) 2329–2334. [5] F.P. Melchels, M.A. Domingos, T.J. Klein, J. Malda, P.J. Bartolo, D.W. Hutmacher, Additive manufacturing of tissues and organs, Prog. Polym. Sci. 37 (2012) 1079–1104. [6] J.-P. Kruth, L. Froyen, J. Van Vaerenbergh, P. Mercelis, M. Rombouts, B. Lauwers, Selective laser melting of iron-based powder, J. Mater. Process. Technol. 149 (2004) 616–622. [7] J. Sun, Y. Yang, D. Wang, Mechanical properties of Ti–6Al–4V octahedral porous material unit formed by selective laser melting, Adv. Mech. Eng. 2012 (2012). [8] B. Vandenbroucke, J.-P. Kruth, Selective laser melting of biocompatible metals for rapid manufacturing of medical parts, Rapid Prototyp. J. 13 (2007) 196–203. [9] M. Van Elsen, F. Al-Bender, J.-P. Kruth, Application of dimensional analysis to selective laser melting, Rapid Prototyp. J. 14 (2008) 15–22. [10] E. Yasa, J.-P. Kruth, Microstructural investigation of selective laser melting 316 L stainless steel parts exposed to laser re-melting, Procedia Eng. 19 (2011) 389–395. [11] D. Jevremovic, T. Puskar, B. Kosec, Đ. Vukelic, I. Budak, S. Aleksandrovic, et al., The analysis of the mechanical properties of F75 Co–Cr alloy for use in selective laser melting (SLM) manufacturing of removable partial dentures (RPD), Metalurgija Zagreb 51 (2012) 171. [12] M. Lindner, S. Hoeges, W. Meiners, K. Wissenbach, R. Smeets, R. Telle, et al., Manufacturing of individual biodegradable bone substitute implants using selective laser melting technique, J. Biomed. Mater. Res. A 97 (2011) 466–471. [13] P. Fox, S. Pogson, C. Sutcliffe, E. Jones, Interface interactions between porous titanium/tantalum coatings, produced by selective laser melting (SLM), on a cobalt–chromium alloy, Surf. Coat. Technol. 202 (2008) 5001–5007. [14] R. Gristina, E. D'Aloia, G.S. Senesi, A. Milella, M. Nardulli, E. Sardella, et al., Increasing cell adhesion on plasma deposited fluorocarbon coatings by changing the surface topography, J. Biomed. Mater. Res. B Appl. Biomater. 88 (2009) 139–149. [15] R.W. Davenport, P. Dou, V. Rehder, S. Kater, A Sensory Role for Neuronal Growth Cone Filopodia, 1993. [16] B.D. Boyan, T.W. Hummert, D.D. Dean, Z. Schwartz, Role of material surfaces in regulating bone and cartilage cell response, Biomaterials 17 (1996) 137–146. [17] L. Qin, Q. Zeng, W. Wang, Y. Zhang, G. Dong, Response of MC3T3-E1 osteoblast cells to the microenvironment produced on Co–Cr–Mo alloy using laser surface texturing, J. Mater. Sci. 49 (2014) 2662–2671. [18] J. Jang, H. Yang, The effect of surface treatment on the performance improvement of carbon fiber/polybenzoxazine composites, J. Mater. Sci. 35 (2000) 2297–2303. [19] Y.N. Podrezov, V. Nazarenko, A. Vdovichenko, V. Danilenko, O. Koryak, Y.I. Evich, Mechanical properties of powder titanium at different production stages. III. Contact formation in powder titanium based on examination of mechanical properties in sintering, Powder Metall. Metal Ceram. 48 (2009) 201–210. [20] S. Eremeev, L.Y. Nemirovich-Danchenko, S. Kul'kova, Effect of oxygen vacancies on adhesion at the Nb/Al2O3 and Ni/ZrO2 interfaces, Phys. Solid State 50 (2008) 543–552. [21] Y. Zhang, P. Bataillon‐Linez, P. Huang, Y. Zhao, Y. Han, M. Traisnel, et al., Surface analyses of micro‐arc oxidized and hydrothermally treated titanium and effect on osteoblast behavior, J. Biomed. Mater. Res. A 68 (2004) 383–391. [22] S.A. Redey, M. Nardin, D. Bernache-Assolant, C. Rey, P. Delannoy, L. Sedel, et al., Behavior of human osteoblastic cells on stoichiometric hydroxyapatite and type A carbonate apatite: role of surface energy, J. Biomed. Mater. Res. 50 (2000) 353–364. [23] K. Das, S. Bose, A. Bandyopadhyay, Surface modifications and cell–materials interactions with anodized Ti, Acta Biomater. 3 (2007) 573–585.

S. Zhang et al. / Materials Science and Engineering C 53 (2015) 50–59 [24] G. Brown, P.J. Hughes, R.H. Michell, Cell differentiation and proliferation—simultaneous but independent? Exp. Cell Res. 291 (2003) 282–288. [25] G. Ryan, A. Pandit, D.P. Apatsidis, Fabrication methods of porous metals for use in orthopaedic applications, Biomaterials 27 (2006) 2651–2670. [26] V. Karageorgiou, D. Kaplan, Porosity of 3D biomaterial scaffolds and osteogenesis, Biomaterials 26 (2005) 5474–5491. [27] S. Van Bael, Y.C. Chai, S. Truscello, M. Moesen, G. Kerckhofs, H. Van Oosterwyck, et al., The effect of pore geometry on the in vitro biological behavior of human periosteum-derived cells seeded on selective laser-melted Ti6Al4V bone scaffolds, Acta Biomater. 8 (2012) 2824–2834. [28] Y. Hoshikawa, E. Yasuda, T. Onoki, M. Akao, Y. Tanabe, Sol–gel derived methylsiloxane–Ca–Nb–Ta coating on titanium surface by double layered structure, Key Eng. Mater. 361 (2008) 693–696. [29] S.A. Pauline, N. Rajendran, Effect of Sr on the bioactivity and corrosion resistance of nanoporous niobium oxide coating for orthopaedic applications, Mater. Sci. Eng. C 36 (2014) 194–205.

59

[30] G. Ramírez, S. Rodil, H. Arzate, S. Muhl, J. Olaya, Niobium based coatings for dental implants, Appl. Surf. Sci. 257 (2011) 2555–2559. [31] S. Lei, M.C. Frank, D.D. Anderson, T.D. Brown, A method to represent heterogeneous materials for rapid prototyping: the Matryoshka approach, Rapid Prototyp. J. 20 (2014) 390–402. [32] K.A. Mumtaz, N. Hopkinson, Laser melting functionally graded composition of Waspaloy® and Zirconia powders, J. Mater. Sci. 42 (2007) 7647–7656. [33] H.I. Medellín-Castillo, J.E.P. Torres, Rapid prototyping and manufacturing: a review of current technologies, ASME 2009 International Mechanical Engineering Congress and Exposition: American Society of Mechanical Engineers 2009, pp. 609–621. [34] H. Seifert, H. Lukas, G. Petzow, Thermodynamic optimization of the Ti–C system, J. Phase Equilib. 17 (1996) 24–35. [35] B.S. Kamber, A. Greig, R. Schoenberg, K.D. Collerson, A refined solution to Earth's hidden niobium: implications for evolution of continental crust and mode of core formation, Precambrian Res. 126 (2003) 289–308.