**1. Introduction**

Improving the biocompatibility of titanium and titanium alloy prostheses is a major challenge for the biomedical industry and has been the subject of much scientific interest in recent years [1,2]. Among all the techniques allowing the improvement of the cells adhesion and proliferation, severe plastic deformation (SPD) of the material has been studied for about one decade and, since then, been the subject of several investigations. SPD makes it possible to refine the microstructure to obtain, under some specific conditions, a nanostructure. The first biocompatibility study on pure titanium deformed by SPD was carried out by Kim et al. in 2007 [3]. The results have shown that the refinement of the microstructure by equal-channel angular pressing (ECAP) leads to a better wettability and therefore a better adhesion and proliferation of fibroblasts after two and five days. In a second study, the same research group has shown that the cytotoxicity remained unchanged after SPD [4]. Other teams have shown that ECAP also leads to a better proliferation of mouse fibroblasts after 72 h [5] and a better adhesion of human mesenchymal stem cells (MSCs) [6]. Further, other investigations have shown that ECAP can be followed by chemical etching or sanding to create surface pores that further increase the bio-integration [7,8]. The high pressure torsion (HPT) [9] is another technique of grain refinement by

core SPD that can improve the biocompatibility as illustrated on NiTi [10] alloys as well as on pure titanium by modification of its protective oxide layer [11]. This modification of the oxide layer on Ti-based materials has recently been demonstrated by other teams using other SPD techniques [12–14]. These positive results on the effect of core SPD have nonetheless been counterbalanced by some studies which have shown no beneficial effect of SPD or equivalent results for coarse-grain and nano-grain microstructures [15,16].

These core SPD techniques impart the plastic deformation to the overall material while, in many cases, only the surface is an important issue in terms of bio-integration. Thus, an interesting prospect for bio-integration is to modify the surface properties while preserving at core the mechanical properties of the raw material. With this respect, techniques have been applied to create surface graded microstructures by surface SPD such as, for example, sliding friction treatment (SFT) [17] as well as surface mechanical attrition treatment (SMAT) [18]. The first one has shown promising results in terms of biocompatibility [13] but remains very difficult to implement industrially. The SMAT technique, also called ultrasonic shot peening (USSP), is already used in industry to increase the fatigue resistance of prosthesis [19] and has already proven its ability to increase the biocompatibility of titanium alloys. SMAT consists of plastically deforming the sample surface by moving shots set in motion by an ultrasonic vibrating part called sonotrode. The main difference compared to conventional directional peening is that, in the case of SMAT, the shots are moving within a sealed chamber giving them random impact trajectories on the surface, enhancing the superficial structural refinement [18,20]. A complete description of the treatment can be found in [21]. This treatment results in nanostructuring of the material but also in an increase in the surface roughness [18,20,21] and, apparently, an increase in surface wettability [22]. Results on pure titanium have revealed an increase in adhesion and viability of MSCs [23]. In-vivo tests on rabbits have been done to compare SMATed and raw Ti6Al4V and the study showed better bio-integration of the surface SPD materials within the bones after four, eight and 12 weeks [24].

The results on bio-integration by SPD nevertheless raise an important question: to greatly improve the biocompatibility, is it better to change the roughness, the size of the microstructure or the chemistry of the material? To answer this question, the latest advances in additive manufacturing that allow to manufacture parts made of functionally graded material (FGM) are here tested in combination with SMAT to ensure that all preparations have been made., Following this, the cell cultures are made under exactly the same atmosphere and in the same environment (temperature, time).

These types of materials, born in the 1980s in Japan [25], can be classified into three distinct groups depending on the nature of the gradients: gradient in microstructure, gradient in porosity and gradient in composition. In the present work, gradients in composition (Ti-Nb and Ti-Mo) will be tested. The functionally graded composition is defined by Pei et al. as "a change in composition across the bulk volume of a material aimed to dynamically mix and vary the ratios of materials within a three dimensional volume to produce a seamless integration of monolithic functional structures with varied properties" [26]. FGM are already used to provide an enhanced substitute for the coating in orthopedic implants, thus avoiding the sudden change in chemical composition and the "peeling-off" effect of the coated [27]. Several advantages can be achieved, such as improving the fixation of implant to bone, enhancing the stress shielding phenomena, hardening the articulating surface, and removing interfacial stresses between the implant and bone.

Bogdanski et al. were the first to use a FGM metal (Ni-Ti) ranging from 0 to 100% to find the composition with the best biocompatibility [28]. Unfortunately, the combination of cold isostating pressing (CIP) + hot isostating pressure (HIP) + vacuum welding processes used at this time was both time-consuming and costly.

Nowadays, direct energy deposition (DED) makes it possible to manufacture FGMs more rapidly and at a lower cost with two or more materials. This additive manufacturing technique, that uses a powder spray, has been successfully tested on FGM titanium-based alloys [29,30]. Since it can be used to manufacture prostheses with surface properties that are different from those of the core, these

FGMs have a strong potential for the future in the biomedical industry [31]. Among all material couples, titanium-niobium alloys and titanium-molybdenum alloys are excellent candidates for new prostheses generation [32,33]. In this interest, the aim of this study is thus to investigate the effect of chemistry, roughness and SMATed microstructure on cell adhesion and proliferation. To this end, FGMs have been made based on the Ti-Nb and Ti6Al4V-Mo constituents and then treated by SMAT. The FGMs compositions range from 100% Ti to 100% Nb as well as 100% Ti6Al4V to 100% Mo and were manufactured with a DED technique called CLAD® [34,35]. The FGMs were investigated in their polished (P), SMATed (S) and SMATed + polished (S+P) conditions then human mesenchymal stem cells were deposited and counted as function of time from three up to 21 days.
