Permanent wettability of a novel, nanoengineered, clinically available, hyaluronan‐coated dental implant

Abstract The objectives of this study are to evaluate long‐term wettability of novel surface‐engineered, clinically available dental implants, featuring a surface nanolayer of covalently linked hyaluronan, and to confirm the relationships between wetting properties and surface nanostructure and microstructure. Wettability measurements were performed on clinically available hyaluronan‐coated Grade 4 titanium implants, packaged and sterile, that is, in the “on the shelf” condition, after 1 year from production. Wetting properties were measured by the Wilhelmy plate method. Analysis of the surface structure and chemistry was perfomed by X‐ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) and energy‐dispersive X‐ray (EDX) analysis, atomic force microscopy (AFM), and ζ‐potential measurement, either on implants or disks or plates subjected to the same surface‐engineering process. Results show that hydrophilicity and ensuing capillary rise of the hyaluronan‐coated implant surface is unaffected by aging and dry storage. Chemical analysis of the implant surface by XPS and evaluation of the ζ potential indicate that hyaluronan chemistry and not that of titanium dictates interfacial properties. Comparison between XPS versus EDX and SEM versus AFM data confirm that the thickness of the hyaluronan surface layer is within the nanometer range. Data show that nanoengineering of the implant surface by linking of the hydrophilic hyaluronan molecule endows tested titanium implants by permanent wettability, without need of wet storage as presently performed to keep long‐term hydrophilic implant surfaces. From an analytical point of view, the introduction in routine clinical practice of nanoengineered implant surfaces requires upgrading of analytical methods to the nanoscale.

escape this trend. Based on in vitro and in vivo evidences of better periimplant bone regeneration on high energy, or hydrophilic, titanium surfaces (Buser et al., 2004;Gittens et al., 2013;Park et al., 2012;Schwarz et al., 2007;Wennerberg et al., 2014), the SLActive concept was introduced in clinical practice (Chambrone, Shibli, Mercúrio, Cardoso, & Preshaw, 2015). The SLActive approach involves packaging of dental implants in saline solution, to preserve the freshly prepared microrough surface from contact with the atmosphere. It is based on knowledge developed by a prominent founding father of biomaterials surface science, Bob Baier, who discussed the concept of storage in water to preserve high-energy surfaces of titanium dental implants in the 1980s and early 1990s (Baier & Meyer, 1988).
From a surface chemico-physical point of view, wettability of titanium surfaces is pretty well understood. Titanium, or the titanium oxide surface layer produced on contact with atmospheric oxygen, belongs, as almost every metal or ceramic materials, to the domain of "high-energy surfaces" (Zisman, 1964). In practice, their internal cohesive energy leads to an excess specific surface free energy some order of magnitude higher as compared to that of water. As a consequence, every pristine metal or ceramic surface, titanium and titanium oxide included, is fully wettable by water, or hydrophilic. The very same high surface energy, however, promotes adsorption of lowenergy ubiquitous hydrocarbons from the atmosphere. In practice, every metal or ceramic surface exposed to the atmosphere becomes hydrocarbon rich, in a time-dependent way, turning from a highenergy to a low-energy surface, whose specific surface free energy is of the same order of magnitude, and in general lower, as compared with that of water. Hence, the finite water contact angles, or hydrophobicity, detected on real-world titanium surfaces. In the case of titanium dental implants, the basic determinant of excess specific surface free energy, that is, the surface chemical composition, combines with the other properties that affect macroscopic wetting behavior and capillary rise, that is surface microtopography and thread size and shape (Rupp, Scheideler, Rehbein, Axmann, & Gels-Gerstorfer, 2004), yielding increasingly hydrophobic surfaces as time from production goes bay. This behavior has been defined "biological aging" (Att et al., 2009;Choi et al., 2017aChoi et al., , 2017b, with a view to its biological effects. Its cause, however, is plainly chemical-physical, as just discussed. The hitherto most investigated chemical-physical approaches to exploit biological responses driven by hydrophilic implant surfaces involve preventing hydrocarbon adsorption by packaging in water (i.e., the already quoted SLActive approach) or getting rid of adsorbed hydrocarbons by UV light or glow discharge cleaning just before usage by side-chair devices (Canullo et al., 2016;Choi et al., 2017aChoi et al., , 2017b. Advances in nanotechnology and surface engineering offer further keys to hydrophilic implant surfaces: Wetting phenomena are exquisitely surface sensitive, that is, they are driven by the chemistry of the outermost atomic or molecular layer (Adamson & Gast, 1997).
Thus, an intrinsically hydrophilic, adsorption-resistant, nanometer-thin, chemically linked surface layer, provided of course it does not adversely affect osteointegration, can impart permanent wettability to titanium implant surfaces, with no need of wet-storage or hydrocarbon removal to restore hydrophilicity. Moreover, due to its nanosize, it does not affect implant surface microtopography and it is not subjected to massive delamination, as reported in the days of micrometer-thick hydroxylapatite coatings. Biomolecular nanoengineering of implant surfaces recently entered clinical practice, through novel dental implants featuring a surface nanolayer of permanently linked hyaluronan, or hyaluronic acid (HY; Morra et al., 2018). The relevance of HY molecular interactions and signaling properties for tissue regeneration and osteointegration, as well as references to relevant in vitro and in vivo works, can be found in the quoted paper. Besides, HY is also a paradigmatic hydrophilic biomolecule, and many of its current applications in medicine are based on its hydration properties (Knopf-Marques et al., 2016;Morra, 2005;Sudha & Rose, 2014). On the light of claimed role of surface hydrophilicity on periimplant tissue response, it is of interest to investigate the effect of the surface-linked hydrophilic HY nanolayer on implant wetting properties.
The scopes of the present work are the following: to evaluate long-term wettability (>1 year on the shelf after production) of a novel surface-engineered, clinically available dental implant, featuring a surface nanolayer of permanently linked HY; to confirm the relationships between wetting properties and surface nanostructure and microstructure; and to highlight the increasing role of finely tuned surfacesensitive analytical techniques and relevant chemical-physical background in the understanding of implant clinical response, on the light of quick advancements of dental implants surface technology towards nanoengineering.

| METHODS AND MATERIALS
Measurements were performed on 4.1 × 16 Shapeone B HYHA implants produced by iRES SAGL, Riva di Caccia, Lugano, Switzerland (ref. code S1B4116HYHA). The macrodesign of the implant is shown in Figure 1. It belongs to the class of "hybrid" implants, meaning that, depending on the implant length, approximately 2/5 of the upper threaded portion is machined and the lower 3/5 portion is sandblasted and acid etched. Moreover, the whole implant surface is further coated by a nanolayer of HY, as described in Morra et al., 2018. Tested implants were chosen in order to evaluate the effect of aging on wetting properties. Thus, implants with approximately 1 year and a few month aging (Lot 20173288 and Lot 20174409, respectively) were taken from the shelf. Implants were ready for clinical use that is packaged and sterile. The following measurements were performed:
Shortly, a microbalance measures the force exerted on a solid (the implant, in the present case) as it is immersed in a liquid (Milliq water in the present experiments). The total force can be written as where W is the weight of the sample (i.e., its mass times the gravitational acceleration), γ L the surface tension of the wetting liquid, p the sample perimeter, and F b the buoyancy. Because W can be measured by the microbalance, F b can be zeroed by extrapolation to zero depth of immersion and p can be measured independently; Equation (1) is commonly used to calculate the contact angle ϑ of a liquid of known surface tension γ L on a solid of known perimeter p, or to measure the surface tension γ L of a liquid that completely wets (ϑ = 0, cos ϑ = 1) a solid of known perimeter p.
In the present case, as widely described in the relevant section, the Wilhelmy plate experiments can be used just for qualitative purposes, for the following reasons: (a) the sample perimeter p is not constant, and it is a function of implant geometry and immersion depth; (b) the solid-liquid wetting interaction is dictated not just by the equilibrium contact angle ϑ (Rupp et al., 2014) but also by the implant surface microtopography, thread geometry, and screw pitch.
Shortly, a dental implant is a highly unideal subject; measurements intended to extract contact angle data should be performed on cylinders or regularly shaped samples. Taking these considerations into account, the Wilhelmy plate experiment is however fit to show the qualitative outcome of wetting interactions and capillary phenomena, as described in Section 3.

| Scanning electron microscopy and energydispersive X-ray analysis
The surface topography of the implants was evaluated by scanning electron microscopy (SEM). Analysis was conducted using an EVO MA 10 SEM (Carl Zeiss Microscopy GmbH, Jena, Germany). The electron acceleration voltage was maintained at 20 kV, the working distance between 10 and 12.5 mm. These parameters are reported in the images, along with the level of magnification (MAG) and the kind of detector utilized (Signal A = SE1 or CZ BSD).
Energy-dispersive X-ray (EDX) analysis was performed using an Aztec microprobe (Oxford Analytica), equipped with an X-act X-ray detector, enabling a resolution better than 129 eV.
2.3 | X-ray photoelectron spectroscopy X-ray photoelectron spectroscopy (XPS) analysis was performed using a Perkin Elmer PHI 5600 ESCA system (PerkinElmer Inc., Waltham, Massachusetts, USA). The instrument is equipped with a monochromatized Al anode operating at 10 kV and 200 W. The diameter of the analyzed spot is approximately 500 μm; the analyzed depth about 5 nm. The base pressure was maintained at 10-8 Pa. The angle between the electron analyzer and the sample surface was 45°.
Analysis was performed by acquiring wide-range survey spectra (0-1,000 eV binding energy) and detailed high-resolution peaks of relevant elements. Quantification of elements was accomplished using the software and sensitivity factors supplied by the manufacturer.
High-resolution C1s peaks were acquired using a pass energy of 11.75 eV and a resolution of 0.100 eV/step.
Beside the just reported evaluation, the following test were made on disks or plates subjected to the same HY coating process.

| Atomic force microscopy
Atomic force microscopy (AFM) was used to explore the surface nanotopography of 5 mm diameter machined Grade 4 Ti disks, both uncoated and HY coated. Measurements were performed using an  Streaming potential and streaming current measurements are known to be useful methods for the investigation of charge displacement in the electrical double layer caused by an external force shifting the liquid phase tangentially against the solid (Luxbacher, 2014).
The ζ-potential value is referred to the approach developed by Helmholtz and Smoluchowski: In this equation, dI str dΔ p coefficient is the streaming current, L is the length of the slit channel formed between the two samples forming the capillary, and A is the area cross section of the capillary. η and ε are the viscosity and dielectric coefficient of the solution forced to pass through the capillary. If we use the Ohm's law I str = U str /R (where R is the electrical resistance inside the streaming channel and U str is the potential) Equation (2) became However, if the solid sample contributes to the conductance inside the streaming channel (i.e., metals), the calculation of streaming potential provides an apparent value of zeta potential, because the measurement of the cell resistance R is affected. Furthermore, the application of Helmholtz and Smoluchowski approach is possible just with planar solids surface, as in our case, since needs an exact knowledge about the geometry of the streaming channel, that is, the cell constant L/A. For the rectangular titanium slit, the length L and width are determined by the solid sample size; instead, the gap height is calculated from the measured volume flow rate of liquid passing through the streaming channel and the generated differential pressure.
All calculations were performed by the instrument software.

| RESULTS
3.1 | Wettability Figure 2 shows macroscopic evidence gathered from Wilhelmy plate measurements of the tested implants. In particular, the image shows a 1.5-s timeframe sequence, that is, the time interval between two subsequent frames is 0.5 s, gathered on Lot 20173288 implants. The implant is hanged to the microbalance through a threaded connector and suspended over a slowly (60 μm/s) rising becker filled with MilliQ water (measured surface tension 71.4 mJ/m 2 ), that is, from the beginning to the end of the timeframe sequence of Figure  The same behavior is qualitatively captured by the outcome of the Wilhelmy plate measurement, that is, force versus distance wetting loops (Rupp et al., 2014). Figure 3 shows the recorded force on two subsequent loops, as follows: the implant touches water at zero depth FIGURE 2 Timeframe image of the Wilhelmy plate experiment performed to evaluate capillary rise. Experimental time is reported in the frames FIGURE 3 Force versus distance wetting loops for the tested implant. ZOI: zero depth of immersion. The implant is first immersed for 5 mm in water (to Point 1), then withdrawn up to Point 2, then immersed again for 7 mm (Point 3), then completely withdrawn of immersion. Immediately, a strong positive force (Equation (1), where W is zeroed by the software and F b is still negligibly small) is exerted on the implant perimeter by the surface tension of water, operating at full force (cos ϑ = 1). The implant is first immersed for 5 mm in water (to Point 1), then withdrawn up to Point 2, then immersed again for 7 mm (Point 3), then completely withdrawn. No hysteresis is observed (i.e., differences between force recorded on advancing and receding, Rupp et al., 2004) because the surface is fully wet and the water front both on advancing and on receding slides over a water layer. Note that, while according to Equation (1), the recorded force is a function of the sample perimeter, hence, it should be higher on thread apex and lower in valleys; no evidence of force modulation by threads is detected in the graph of Figure 3. This is because threads are filled by water draw by capillary rise and the water front "feels" an apparent constant perimeter. Further note, on implant withdrawal, the remarkable increase of weight (double-headed arrow, Point 4 in Figure 3) due to water captured within threads and surface microroughness.

Altogether, both Figures 2 and 3 indicate a totally hydrophilic behavior
and no evidence of aging on dry storage. The surface chemico-physical basis behind these results is investigated in the forthcoming sections. hence the failure of EDX, whose sampling depth is a few micrometers, that is, three order of magnitude thicker, to clearly detect it.

| X-ray photoelectron spectroscopy
Chemical composition data obtained on probing the surface of tested implants by XPS is reported in Table 1 stoichiometry can be inferred from obtained data. The resulting information is that an organic overlayer, whose thickness is bigger than the XPS sampling depth (about 5 nm), homogenously covers the underlying implant.
High resolution C1s, O1s, N1s peaks were obtained to gather more information on the molecular structure of the surface layer.

| Atomic force microscopy
Supplementary information on the surface structure were obtained by analysis of machined disks uncoated and subjected to the same surface modification treatment as implants. Main results from noncontact AFM analysis are reported in Figures 6 and 7. Figure 6a shows a 10 × 10-μm field of view obtained on the machined surface (not coated). The imaged area has a similar size as compared with that shown in Figure 4a (15 × 10 μm). In Figure 6a, the main features are the already discussed parallel grooves due to machining. Moving to Figure 6b, that is, to the coated surface, thus the same subject shown in Figure 4a, a significantly different picture is observed. While the presence of the parallel grooves can still be inferred, a submicrometer bumped structure can be clearly detected. This is even more evident in Figure 7, comparing 5 × 5-μm fields of view of the machined (a) and coated (b) surface (that is, to be compared with the SEM view of Figure 4b). In agreement with XPS findings, the nanotextured structure due to the coating covers homogenously the surface, that is, the underlying Ti does not "show through" (else, some Ti would be detected in the XPS spectra). Note that the vertical scale of the image is of the order of some tens of nanometers. Figure 8 is a height profile obtained by drawing a 10-μm line in Figure 6b. Bumps range from a few nanometers to a few tens of nanometers. It is actually the exceptional vertical resolution of the AFM technique that allows to imagine topography details of the HY overlayer through nanoroughness contrast.

| ζ-potential measurement
Eletcrokinetic effects at the solid-liquid interface were evaluated through ζ-potential measurement. In particular, as described in Section 2, the streaming current was measured as a function of pH, in the 2.8 to 8.2 range, in 1-mM KCl solution.
Results are shown in Figure 9. In the case of Ti, the pH scan yields a curve that moves from positive ζ-potential values at low pH to negative pH values at high pH, crossing the isoelectric point at 3.67. Hence, Ti is negatively charged at physiological pH. This behavior is qualitatively consistent with results obtained by Roessler, Zimmermann, Scharnweber, Werner, and Worch (2002)   Lot 20174409 microrough portion 32.7 ± 0.5 8.5 ± 0.3 58.8 ± 0.5

FIGURE 5
High-resolution C1s peak of the tested implant surface obtained by XPS analysis. The peak is curve fitted by four components: 1 = C-C, C-H; 2 = C-N; 3 = C-O; 4 = C¼O, N-C¼O, and O-C¼O following Shard et al. (1997)    repair (Abatangelo & Weigel, 2000;Evered & Whelan, 1989;Kennedy, Phillips, Williams, & Hascall, 2002;Laurent, 1998). There is a significant background on its use in the surface modification of medical materials and devices (Morra, 2005). In the spirit of biomolecular modification of implant surfaces (Puleo & Nanci, 1999), the presentation of HY molecular motifs at the implant/tissue interface could promote healing mechanisms not provided by the Ti interfacial chemistry.
Beside specific interactions, HY is also a paradigmatic hydrophilic molecule (Toole, 2000). The present work shows the impact of surface engineering by HY on wetting properties of Ti dental implants. The wetting behavior of the basic device of dental implantology, the titanium fixture, is bounded by native and induced aspects. The most important native feature is the chemical nature of Ti that leads to a high-energy titanium oxide surface, as described in Section 1. Induced aspects involve the hierarchical structure of threads, screw pitch, concave grooves generated by sandblasting (in the range of 5 to about 70 μm), and pores of different shapes created by acid etching (in the range of several hundred nanometers up to 3 μm). This combination naturally leads to hydrocarbon adsorption (Att et al., 2009;Choi et al., 2017aChoi et al., , 2017b) and a wetting behavior that, as explained in basic teaching of wettability science (Cassie & Baxter, 1944, Dettre & Johnson Jr, 1964, Wenzel, 1936a, 1936b, is magnified by topographic details, leading to hydrophobic implant surfaces and large contact angle hysteresis on aging, as nicely described by Rupp et al. (2014). Within these boundaries, prevention of hydrocarbon adsorption by wet storage (Baier & Meyer, 1988) or removal of adsorbed hydrocarbons by discharge techniques just before usage (Canullo et al., 2016;Choi et al., 2017aChoi et al., , 2017b stop or reverse the natural trend and provide highenergy surfaces whose hydrophilicity is further enhanced by capillary wicking into grooves and pores. surfaces; hence, it does not promote hydrocarbon adsorption, and it keeps its chemical identity on aging ( Figure 5). It presents at the solid/liquid interface a high density of hydrophilic moieties, that is, polar carbon-oxygen bonds, hydroxyl groups, and carboxylate functionalities ( Figure 5 and Table 2). The latter control charge behavior at the solid/liquid interface as shown in Figure (Buser et al., 2004;Gittens et al., 2013;Morra et al., 2009;Park et al., 2012;Schwarz et al., 2007;Wennerberg et al., 2014) and provide the ethical basis to put all the instruments of surface science at the service of dental implantology.

| CONCLUSIONS
In conclusion, data presented in this work show that nanoengineering