# Fileset

[132_24090.pdf](https://mdr.nims.go.jp/filesets/cfff3364-585f-44cf-9b98-d1769d5f806b/download)

## Creator

Masami Hashimoto, Norio Yamaguchi, Soma Hashimoto, Hidenobu Murata, Satoshi Kitaoka, Daisaku Yokoe, Taishi Ito, Takeharu Kato, Hiroyasu Kanetaka, [Hideki Kakisawa](https://orcid.org/0000-0002-7448-0989)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial properties](https://mdr.nims.go.jp/datasets/8131cca7-c0ae-4f0c-8587-d86bcef0383e)

## Fulltext

Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial propertiesFULL PAPERDevelopment of segmented titanium oxynitride layerwith spinform surfaces exhibiting antibacterial propertiesMasami Hashimoto1,³, Norio Yamaguchi1, Soma Hashimoto1, Hidenobu Murata1, Satoshi Kitaoka1,Daisaku Yokoe2, Taishi Ito2, Takeharu Kato2, Hiroyasu Kanetaka3 and Hideki Kakisawa41Materials Research and Development Laboratory, Japan Fine Ceramics Center, 2–4–1 Mutsuno, Atsuta-ku, Nagoya 456–8587, Japan2Nanostructures Research Laboratory, Japan Fine Ceramics Center, 2–4–1 Mutsuno, Atsuta-ku, Nagoya 456–8587, Japan3Graduate School of Dentistry, Tohoku University, 4–1 Seiryo-machi, Aoba-ku, Sendai 980–8575, Japan4National Institute for Materials Science, 1–2–1 Sengen, Tsukuba, Ibaraki 305–0047, JapanWe have developed the titanium oxynitride (TiON) layer which exhibits antibacterial activity againstEscherichia coli without sunlight irradiation. This TiON layer with the segmented structure was formed on apre-oxidized Ti substrate by the electron beam physical vapor deposition method while spraying the substrateand Ti target with N2-0.1% O2 gas. An XRD pattern of the surface of the segmented layer revealed that the layerhas a rock salt structure, which is preferentially deposited in a [111] direction to the substrate surface. TEMobservation of the TiON layer showed that the columnar segments have a feather-like structure. The width of thesegments near the surface was 100–500 nm, and their outermost surfaces were covered with nanometer-sizedspines of 10–50 nm width. The TiON layer contained about 20 at% nitrogen as determined by XPS, and thesurface was slightly negatively charged with a zeta potential of ¹0.9mV at pH 7.2. The maximum averagepressure in the elastic deformation region for the layer was over 100MPa, which was much higher than theaverage occlusal pressure (<10MPa). It is concluded that the segmental structure of the film facilitates the elasticdeformation without distorting of the unique surface structure in response to external stress.Key-words : Segmented TiON layer, Spinform, Electron beam physical vapor deposition, Antibacterial property,Nanoindentation[Received August 8, 2024; Accepted September 26, 2024; Published online October 24, 2024]1. IntroductionTi and its alloys have extensive experience as a dentalimplant or an artificial hip joint because of its excellentcorrosion resistance and biocompatibility due to the for-mation of a strong passive layer. Implant surfaces arerequired not only to bond to the alveolar bone, but also tohave antibacterial properties, and other functions.As a method to impart antibacterial properties to Ti andTi alloys, immobilization of ions such as copper,1) sil-ver2–5) or iodine6,7) in the surface oxide layer is used toimpart antibacterial properties to Ti implants, etc., and acontrolled slow release of these ions, that is safe for livingorganisms is required. It has been reported that the anti-bacterial property is due to the release of these ions. Whenthe antibacterial property is expressed by elution of ions, itis important to control the amount of immobilized ions andthe elution rate.There is also a method to immobilize a photocatalyti-cally active film on Ti and Ti alloys to develop antibac-terial properties. First, pure titanium is heat-treated in anammonia atmosphere to form a nitrogen-incorporatedtitanium dioxide layer that exhibits antibacterial propertiesin response to visible light.8) Second, TiON sputtered poly-ester surfaces are activated by sunlight irradiation, result-ing in accelerated bacterial inactivation within minutes.The absorption in Kubelka–Munk units of the TiON filmwas observed to be directly proportional to the time ofinactivation of Escherichia coli as determined by diffusereflection spectroscopy.9) These methods require exposureto light to develop antibacterial properties.Recently, focusing on the fact that insect wings have astrong bactericidal effect, it was found that mimicking theirsurface shape can produce a very high bactericidal ef-fect.10–17) This function is expressed by transforming thesurface of the component into a structure with a forest ofspines of the order of several tens of nanometers in width,which rupture the cell membranes of bacteria attached tothe surface of the spines. If the nanostructure can be simu-lated on the surface of an implant, it is expected to producea revolutionary antibacterial property that is not dependenton drugs or other agents.However, a load of approximately 100 kN is typicallyapplied during tooth occlusion, and there is concern thatsuch a ceramic layer with thin spines may be easily³ Corresponding author: M. Hashimoto; E-mail: masami@jfcc.or.jpJournal of the Ceramic Society of Japan 132 [12] 681-689 2024DOI https://doi.org/10.2109/jcersj2.24090 JCS-Japan©2024 The Ceramic Society of Japan 681This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/),which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.https://doi.org/10.2109/jcersj2.24090https://creativecommons.org/licenses/by/4.0/worn away by stress concentration caused by repeatedocclusion.Here, we focused on the electron beam physical vapordeposition (EB-PVD) technique, which enables coatingwith a unique surface structure.18,19) For example, a targetsource of Y2O3 solid solution ZrO2 (YSZ) is irradiated andthe tip part is molten by an electron beam, and then a YSZvapor deposited film is formed on a substrate placed at apredetermined position. The microstructure of the seg-ments, such as the crystal orientation, void amounts, andspacings, and segment surface morphology, can be con-trolled by adjusting the deposition conditions, such aselectron beam power, heating temperature and substraterotation speed. This not only leads to the expectation thatnanoscale spine surfaces will be formed, but also that theporous segments will exhibit deformation tolerance toexternal stress due to their low apparent Young’s modulus.In other words, the destruction of the surface morphology,which has a spine structure with antibacterial properties,can be suppressed by relieving the stress concentrationthrough local deformation of the contact area due to toothocclusion.We attempted to form a biocompatible TiO2-based filmwith a spine structure on the surface of Ti, but considered acomposition in which N was introduced to suppress diffu-sion and facilitate the formation of the spine structure.In this study, we first explored the possibility of creatingsurface modified layers of TiON segments with spine sur-faces of tens of nanometers on Ti substrates using EB-PVD, and then evaluated the antibacterial property withoutsunlight irradiation and the deformation tolerance of thesegments to external stress.2. Experimental procedure2.1 Formation of TiON segmented layer byEB-PVDA substrate for coating was prepared from high-purityTi (Ti = 99.5%, TI-452664, Nilaco Co., Tokyo, Japan),which is used as a model material for Ti implant alloys.After the substrate was cut to ¯10 © 3mm, the coatingsurface was polished to a mirror finish. In this study, toimprove the adhesion between the Ti substrate and theTiON film, the mirror-polished substrate was pre-oxidizedat 700 °C for 1 h in a N2 atmosphere (PO2 = 10¹17 Pa) withPO2 adjusted by an oxygen pump according to the methoddescribed in Ref. 23) to form a rutile-type TiO2 layer witha thickness of 70 nm on the substrate. This layer showedan exceptional ability to form apatite in artificial bodyfluids and to induce osteoblasts.The pre-oxidized substrates were then mounted on a Tijig in a chamber of an EB-PVD system with a semicon-ductor laser heating system. High-purity Ti (Ti = 99.5%,¯40 © 100mm, TI-452791, Nilaco Co., Tokyo, Japan) ofthe same grade as the substrates was used as the Ti targetsource. The distance between the target and the substrateswas 600mm. The substrates and the target were sprayedwith N2 gas containing a small amount of O2 gas (N2-0.1% O2) for sample A and O2 gas for sample B at a flowrate of 100 cc/min. The fixture holding the substrates wasrotated at 30 rpm during coating with a 12 kW electronbeam power applied to the target. By rotating the substrateat this time, the deposition surface is periodically shad-owed against the highly directional flow of the materialvapor (shadowing effect), resulting in the formation ofsegments with a longitudinal crack structure in the thick-ness direction. The coating surfaces were heated to ap-proximately 550 °C by laser irradiation and radiative heattransfer from the molten pool of the target. The total pres-sure in the chamber during coating was in the order of0.1 Pa.2.2 Surface structure analysisThe surface morphology of the samples was observedby scanning electron microscopy (SEM; SU8000, HitachiHigh-Tech., Tokyo, Japan). The crystalline phases of thesurface layers formed by the pre-oxidation and the sub-sequent coating were characterized by powder X-ray dif-fractometry with Bragg-Brentano geometry (SmartLab,Rigaku Corporation, Tokyo, Japan) with CuK¡ radiationat 40 kV and 50mA. Incident slit was 1° and detector wasD/tex Ultra 250. The surfaces were etched by Ar ion(4 kV) for 4min at sputtering rate of 0.01 nm/min, andthen the composition of the etched surface layers was eval-uated by X-ray photoelectron spectroscopy (XPS; PHI X-tool, ULVAC-PHI, Kanagawa, Japan). The X-ray sourcewas monochromatic AlK¡ radiation at 14 kV and 25mA.The binding energy was calibrated using the C 1s photo-electron peak of contamination carbon at 284.8 eV as a ref-erence. XPS peak analysis was performed, Gauss-Shirleybackground was subtracted from all spectra before fitting.2.3 Surface potential measurementThe zeta potential was measured under wet conditions,which corresponded to the electrostatic potential in theliquid medium. The Ti plates (¯10 © 3mm) were preparedfor zeta potential measurement as described in section 2.1.The pre-oxidized Ti and the surface coated by EB-PVDwere electrically grounded to allow any stray charge to dis-sipate, and immediately placed in a zeta potential measure-ment apparatus (ELSZneo; Otsuka Electronics Co., Shiga,Japan) using a glass cell for the plate sample. The zetapotentials of three samples were measured in a 10mMNaCl solution at pH 7.2.2.4 Cross-sectional analysis of the segment-ed TiON layerThe Pt thin films on the outermost surfaces of the cross-sectional images were coated to prevent damage to theedges during cross-sectional processing with an ion mill-ing device. The TiON layer formed on the pre-oxidized Tisubstrate was thinned in a Hitachi NB5000 focused ionbeam (FIB) scanning electron microscopy system at anaccelerating voltage of 40–2 kV. Then, Ar ion beams withan accelerating voltage of 1–0.1 kV were then applied tothe thinned specimen to remove the FIB-damaged layerson it using a Gatan PIPS II. Nanostructural character-Hashimoto et al.: Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial propertiesJCS-Japan682ization of the TiON layer was performed using a conven-tional transmission electron microscope (TEM) (JEM-F200, JEOL, Tokyo, Japan) equipped with dual EDSdetectors having a total solid angle of 1.7 sr. Quantitativechemical composition maps of characteristic X-rays of Ti-K¡, N-K¡ and O-K¡ were obtained using the EDS sys-tem. Each pixel of these composition maps has the quan-titative values of Ti-K¡, N-K¡ and O-K¡, respectively.Ti/(N + O) and O/(N + O) value maps corresponding thecompositional maps were drawn using Python (to studythe distribution of N and O in the TiON).20)2.5 Evaluation of antibacterial activityAs a preliminary test, a qualitative test for antibacterialsusceptibility test was performed with n = 2. A nutrientagar was used in Petri dishes (Falconμ plastic dish forgeneral bacteria, Corning Inc., New York, NY, USA) in 15mL aliquots. Physiological saline was prepared by dissolv-ing 8.5 g of sodium chloride (NaCl, Nacalai Tesque, Inc.,Kyoto, Japan) in 1 L of ultrapure water, which was usedafter sterilization at 121 °C for 20min in a high-pressuresteam sterilizer. Escherichia coli (E. coli, JCM5491) wasused as the test bacterial assay. It was used after beingcultured on the nutrient agar medium at 37 °C for 24 h. Thebacterial mass of the cultured E. coli was collected with aplatinum loop and dispersed in physiological saline to pre-pare a stock bacterial suspension (= 106CFUmL¹1). Thebacterial suspension was seeded on each substrate (Ti, pre-oxidized Ti, sample A and sample B), followed by adhe-sion of the film and incubation at 37 °C for 24 h. Afterincubation, the samples were stained with acridine orange,a viable stain, and incubated at 37 °C for 1 h in the dark.After gentle rinsing with sterile PBS to remove excessstaining agent, the live (green) bacteria on the substrateswere examined under a fluorescence microscope (CKX53;Olympus, Tokyo, Japan).2.6 Deformation property of filmsApparent Young’s modulus, penetration depth, andaverage pressure were evaluated within an elastic defor-mation range during load application using the nano-indenter (TI 950 Triboindenter, Hysitron, MN, USA). Thespherical indenter was used with a radius of 9.85¯m. Theelastic deformation region satisfies Sneddon’s expression(1),21,22) where Pm is the mean pressure, P is the appliedload, R is the radius of the spherical indenter, h is thepenetration depth of the indenter, and E is the apparentYoung’s modulus of the film.Pm ¼ P³Rh¼ 4E3³hR� �0:5ð1ÞSince E remains constant in the elastic deformation, theregion where (3P)/(4R0.5h1.5) remains constant as a func-tion of h is assumed to be in the elastic deformation region.Therefore, the average value of the E and the correspond-ing Pm were determined at the maximum values of P and hin this region. The contact radius of indenter, rc, is definedas rc = Rh0.5 for field swan approximation (hc = h/2,where hc is contact depth).3. Results and discussion3.1 Films formed by EB-PVDFigure 1 shows the XRD patterns of the surfaces (a)before and (b) after the pre-oxidation of the Ti substrates,and the film surfaces of the samples (c) A and (d) Bformed by the EB-PVD. Rutile-type TiO2 was formed bythe pre-oxidation of the substrates. For sample A as shownin Fig. 1(c), the diffraction peaks assigned to Ti phasebecomes significantly smaller, and strong diffraction peaksidentified to TiON with a rock salt structure (PDF#04-002-0430) were detected. This film is strongly oriented in thedirection of the [111] plane. On the other hand, for sam-ple B in Fig. 1(d), rutile-type TiO2 phase with a very weakdiffraction peak was detected instead of the peaks assignedto TiON.Figure 2 shows the surface and cross-sectional SEMimages of samples A and B. Sample A has segments about2¯m thick with a vertical crack structure, and the width ofeach segment is several 100 nm, and the surface is com-posed of sharp edges. In contrast, sample B has a film witha thickness of about 1¯m was formed, and its surface wasrelatively smooth. In addition, the segment structure is notclear compared with that of sample A, and there are inho-mogeneous vertical cracks in the direction of the filmthickness at intervals of several micrometers, as shown bythe arrows in the surface image.Figures 3(a), 3(b), and 3(c) show the cross-sectionalTEM image of sample A and the electron diffraction pat-terns of insets (i) and (ii) in the image, respectively. Thenear-surface carbon and W layers were coated to preventedge damage during the cross-sectional milling of thesample. Both (i) and (ii) in the insets refer to TiON. TheTiON segments show a feather-like structure with a pyra-midal tip and their surfaces covered with nano-spines.20 30 60 7040 502θ (deg.)Intensity (a.u.) TiON: TiO2  : Ti200220111(b)(a)(c)(d)Fig. 1. X-ray diffraction patterns of the surfaces of (a) beforeand (b) after the pre-oxidation of the Ti substrates, and the filmsurfaces of the samples (c) A and (d) B formed by the EB-PVD.Journal of the Ceramic Society of Japan 132 [12] 681-689 2024 JCS-Japan683The widths of the segments increase toward the surface.The widths of the top of the segments measured in theimage in Fig. 3(a) are between 100 and 500 nm, and thoseof the spines are between 10 and 50 nm. The film consistsof a TiN layer, a TiO layer, and a TiON layer in order fromthe side of the Ti substrate. The TiN layer progresses dueto the nitridation reaction at the Ti–TiO2 scale interfacewhen the Ti substrate is pre-oxidized in an N2 environmentunder low PO2 conditions at 973K. It is thermodynami-cally formed with a PO2 of 10¹23 Pa or lower.23)When a film is deposited by the EB-PVD method, thesurface temperature of the film to be deposited is generallyexposed to high temperatures of 500 °C or higher due tothe radiant heat associated with the melting of the surfaceof the target source by electron beam irradiation. The con-cern here is that the spines on the surface may disappeardue to accelerated surface diffusion on the film duringhigh-temperature deposition, resulting in a smooth surface.But covalent bonding characteristics in a Ti–O film, whichis expected to have biocompatibility, can be strengthenedby N-doping in the film, resulting in the formation of thespine surface.Figure 4 shows the high-resolution TEM image of thetip of the segment of sample A. The lattice spacing of theTiON layer is 0.245 nm, which corresponds to the latticespacing of the (111) plane of TiON (ICSD426340) withrock salt structure. The TiON crystal is oriented in a direc-tion of the [111] plane relative to the substrate, which isconsistent with the XRD pattern of Fig. 1(c). The surfaceenergy of TiON is highest in the (111) plane, followed bythe (110) and (100) planes, in that order. The (111) plane isunstable because it is composed of the same type of atomsand the surface is highly charged. The (100) layer is themost stable layer because the Ti, N, and O atoms alter-nate like a checkerboard, and their positive and negativecharges cancel each other out. The (110) plane, with alter-nating rows of Ti atoms and rows of O and N atoms, iselectrically neutral but not as stable as the (100) plane. TheTiON layer produced by EB-PVD grows preferentiallytowards the (111) plane, which has a high surface energy,Sample A Sample BGas N2 - 0.1%O2 O2SurfaceCross-section500 nm500 nm1 μm 500 nmPtPtTi TiFilm FilmFig. 2. SEM photographs of the surface and cross-section of the samples A and B.(a) (b) (c)Fig. 3. Cross-sectional TEM image of the sample A (a). (b) and (c) are the electron diffraction patternscollected from the region marked with a white circle in the TEM image (i) and (ii), respectively. Insets (i) and (ii)show the TiON.Hashimoto et al.: Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial propertiesJCS-Japan684and also forms the (100) and (110) planes, which have alow surface energy, and are thought to form a sharp-tippedsurface.Figure 5 shows the bright field STEM image and thecorresponding elemental ratio maps using the EDS ele-mental distribution data of the cross section of sample A;(a) STEM image, (b) Ti/(N + O), (c) vertical average ofTi/(N + O) ratio, (d) O/(N + O) and (e) vertical averageof O/(N + O) ratio. The film consists of a TiN layer, a TiOlayer, and a TiON layer in order from the side of the Tisubstrate. As shown in Figs. 5(b) and 5(d), the surface ofthe TiON segments is oxygen rich compared to the inte-rior, and the interface between the segments and the sub-strate is nitrogen rich. In addition, the Ti/(N + O) rationear the surfaces of the segments is about 0.8, which isoxygen-rich compared to 1.0 for the insides. When a Tisubstrate was preoxidized at 700 °C in N2 gas with anextremely low O2 concentration controlled by an oxygenpump, a layer with a rutile structure with a only 70 nmthick was formed on the substrate as shown in Fig. 1(b),and a TiN layer was also formed just below the scale.23)Therefore, the nitrogen-rich layer just above the substratein Fig. 5 is considered a TiN layer. However, the thick-ness of the O-rich layer in Fig. 5 is sufficiently greater thanthat of the rutile scale formed by the pre-oxidation. Theoxygen-rich segments are probably formed, because theoxygen partial pressure became locally high near the coat-ing surface due to the reduction of the oxide in the initialstage of the segment deposition at 550 °C.The surfaces of samples A and B were each analyzed byXPS, and the compositions were analyzed based on thebinding energy values obtained from the photoelectronspectra. Figures 6(a) and 6(b) show the Ti2p and N1s XPSspectra of samples A and B, respectively. Only sample Acontained nitrogen, and the composition of the layer wasTi45.2O31.8N23.0. Ti/(O + N) is 0.82, which means that thesurface of sample A was Ti poor structure, which is con-sistent with the EDS result as shown in Fig. 5(c).In the case of Ti2p3/2 level of sample A, the bindingenergy of TiN, TiON and TiO2 were determined to be454.5, 456.3 and 458.3 eV, respectively.9) The fractions ofTiN, TiON and TiO2 determined from the deconvolutedFig. 4. High-resolution TEM image of the tip of the segment of sample A.Fig. 5. Bright-field STEM image and the corresponding elemental ratio maps using EDS elemental distributiondata of the cross-section of the sample A; (a) STEM image, (b) Ti/(N + O), (c) vertical average of Ti/(N + O)ratio, (d) O/(N + O) and (e) vertical average of O/(N + O) ratio.Journal of the Ceramic Society of Japan 132 [12] 681-689 2024 JCS-Japan685spectrum were 40.8, 14.3 and 11.5 area%, respectively.The presence of N in the sample A was characterized byusing XPS, the major Ti–N bond in TiON and the minorchemisorbed N–O bond in N1s core levels were located at397.4 and 401.7 eV, respectively as shown in Fig. 6(a).Therefore, the oxygen was incorporated into the rock saltTiN structure during the deposition process under N2-0.1%O2 gas flow. Furthermore, it suggests the presence oftrace amounts of TiO2 near the surfaces of the segments.On the other hand, the composition of the layer of sam-ple B was Ti33.7O66.3. In the case of Ti2p3/2 layer of sam-ple B, the binding energy of TiO, Ti2O3 and TiO2 werelocated at 454.5, 456.1 and 458.3 eV, respectively.24) Thefractions of TiO, Ti2O3 and TiO2 determined from thedeconvoluted spectrum were 35.7, 36.0 and 28.3 area%,respectively.The average zeta potentials at pH 7.2 for samples A andB were ¹3.7 « 1.1mV and ¹23.5 « 0.4mV, respectively.The fact that the zeta potential of sample B was morenegative than that of sample A suggests that the surface ofsample B is covered with a greater number of negativelycharged oxygen ions.3.2 Antibacterial propertyAfter 24 h of the incubation, fluorescence microscopyimages showed that (a) the Ti substrate, (b) the pre-oxidized substrate, and (d) the sample B are covered withnumerous and scattered viable bacteria as shown in Fig. 7.In contrast, a few live bacteria were observed on sample Aas shown in Fig. 7(c). As for the reason why sample Adeveloped antibacterial properties, firstly, the zeta potentialof the segmented TiON membrane is slightly negativelycharged (¹0.94 « 1.62 eV), which is believed to cause anelectrical repulsion between it and the negatively chargedbacteria. However, the zeta potential of pre-oxidized Tiand sample B were ¹6.6 « 0.25 and ¹6.45 « 1.36 eV,respectively. On sample A, it is unlikely that bacteriaadhesion is reduced due to electrostatic repulsion betweenthe substrate and bacteria. Possible denaturation mecha-nisms of bacterial proteins related to antimicrobials includeoxidation by the Mars-van-Krevelen mechanism,25,26) S–Sbond cleavage,27) acid-base reactions,28) and hydrogenbond cleavage.29) The TiON film does not contain Mg, Feor Cu elements that affect protein denaturation. There arereports that the number of bacteria on TiON9) and TiN30)with a smooth surface is reduced by about 50 to 60% evenin the dark. However, as shown in Fig. 7(c), almost no livebacteria are observed on spiny TiON, so there may be acause for the antibacterial properties that does not appearon smooth surfaces. Therefore, another factor may be in-volved in the antibacterial susceptibility.The TiON layer consists of 100–500 nm segments with10–50 nm spines on the surface [Fig. 3(a)]. Segmentalstructured surfaces produced by the EB-PVD process havespikes on the surface that are equivalent in size to thoseeffective for antibacterial properties. In order to kill bac-teria that come in contact with the surface, research is un-derway to control the surface topography (physical geom-etry, nanostructure, nanotexture) of Ti.31–34) The nano-structured surfaces have earned their reputation as mecha-nobactericidal surfaces, independent of chemical effects,because the functionality (ability to kill bacteria) has beenshown to persist across different materials. Thus, webelieve that when bacteria adhere to the spicule, their cellmembranes are destroyed and killed, thus inhibiting bac-terial growth and causing the development of antibacterialproperties.Binding energy (eV) Binding energy (eV)392394396398400402404Binding energy (eV)Intensity (cps)392394396398400402404TiONTiNIntensity (cps)Binding energy (eV)N1s N1sTiN(a) (b)450452454456458460462464466468470TiO22p1/2Intensity (cps)Ti2pTiO22p3/2TiO2p3/2TiO2p1/2Ti2O32p3/2Ti2O32p1/2450452454456458460462464466468470Intensity (cps)TiN2p3/2TiN2p1/2 TiON2p3/2TiO22p3/2TiO22p1/2TiON2p1/2Ti2pFig. 6. Ti2p and N1s XPS spectra of the surfaces of samples (a) A and (b) B.Hashimoto et al.: Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial propertiesJCS-Japan6863.3 Deformation property of filmsThe nano-spiked segments were found to have anti-bacterial properties. In order to understand the deforma-tion tolerance of the segment under occlusal conditions,the deformation characteristics of the elastic deformationregion were investigated using the nanoindentation tech-nique with the spherical indenter. Figure 8(a) shows thetypical indentation load P versus penetration depth hcurves for samples A and B. The maximum applied load(Pmax) was set to 1,000¯N. The maximum penetrationdepth (hmax) at the Pmax for sample A is greater than thatfor sample B. The reproducibility of the p-h curve for sam-ple A is good, but for sample B it varies widely. This ispresumed to be because the through cracks exist unevenlyin the film thickness direction of sample B. Assuming thatSneddon’s equation21,22) is satisfied within the elastic de-formation range of the films, the apparent Young’s mod-ulus (E) for elastic deformation and the maximum defor-mation depth (he-max) within the deformation range areobtained. Figure 8(b) shows the relationship between(3P)/(4R0.5h1.5) and h for each curve in Fig. 8(a). The Eand he-max (arrows in the figure) for each curve are deter-mined from the region where a value on the longitudinalaxis in Fig. 8(b) is constant. The length of the flat area inFig. 8(b), sample A has a longer elastic deformation range.Figure 9 shows (a) E, (b) he-max, and (c) average pres-sure at he-max within the elastic deformation region ofsamples A and B. Each value is the average of three data.The E value of sample A is 1/4 of that of sample B, andthe he-max of sample A is larger than that of sample B.Since the average maximum indentation depth he-max of theindenter in the elastic deformation region of sample A is23 nm, the indenter contact radius rc is 475 nm and theindenter contact area S is 0.708¯m2. Assuming that thesegment width (diameter) is 150 nm, the number of seg-ments in the above contact area above is 40. On the otherhand, since the average he-max of sample B is 12 nm, the rcis 343 nm and the S is 0.369¯m2. As previously men-tioned, the variability in the load-depth curves of sampleA was smaller than that of sample B, and the apparent(a) (b)(c) (d)200 μm 200 μm200 μm 200 μmFig. 7. Viable bacteria stained with live assay on (a) before, (b) after the pre-oxidation of the Ti substrates,(c) sample A and (d) sample B (n = 2).Journal of the Ceramic Society of Japan 132 [12] 681-689 2024 JCS-Japan687Young’s modulus of the sample Awas lower with a longerelastic deformation range. In the case of sample A, eachsegment deforms slightly in response to the indentation bythe indenter, which reduces the stress concentration at theindenter contact area, allowing the entire film to undergosignificant deformation. On the other hand, in sample B,where the segments were thicker and less deformable, thestress is more likely to be concentrated at the indentercontact area, making it more susceptible to the influence ofthe segment shape at that area.4. ConclusionsThe electron beam PVD method deposited TiON layeroriented in the [111] direction on the pre-oxidized Ti sub-strate by heating at 550 °C while blowing N2-0.1%O2. TheN content in the TiON layer was about 20 at%, and thewidth of the top surface of the columnar segments con-stituting the layer was 100 to 500 nm. The width of thespines ranged from 10 to 50 nm. The TiON layers showedantibacterial activity against E. coli. This may be mainlydue to the disruption of the cell membrane by the spikystructure. The TiON layer is easily deformed by externalstress. The average pressure at maximum load in the elas-tic deformation zone of the layer is much higher than theaverage occlusion pressure. In the future, it will be nec-essary to develop TiON films that are not only antibacte-rial, but also biocompatible.Acknowledgments This research was supported byJSPS Grant-in-Aid for Scientific Research 21H01644 and19H05792. We would like to thank Drs. Hideyuki Murakamiand Kimiyoshi Naito, NIMS, and Dr. Takashi Akatsu, TokyoUniversity of Technology, for their cooperation and usefuladvice on the measurement and analysis methods of nano-indentation.References1) K. Suzuki, T. Yokoi, M. Iwatsu, M. Furuya, K. Yokota,T. Mokudai, H. Kanetaka and M. Kawashita, J. AsianCeram. Soc., 9, 1448–1456 (2021).2) K. Suzuki, M. Iwatsu, T. Mokudai, M. Furuya, K.Yokota, H. Kanetaka, M. Shimabukuro, T. Yokoi andM. Kawashita, Molecules, 28, 650 (2023).3) A. Skovager, K. Whitehead, D. Wickens, J. Verran, H.Ingmer and N. Arneborg, Colloid. Surface. B, 109, 190–196 (2013).4) Y. Okuzu, S. Fujibayashi, S. Yamaguchi, K. Masamoto,B. Otsuki, K. Goto, T. Kawai, T. Shimizu, K. Morizane,T. Kawata, Y. Shimizu, M. Hayashi and S. Matsuda,J. Biomater. Appl., 35, 670–680 (2021).5) K. Masamoto, S. Fujibayashi, S. Yamaguchi, B. Otsuki,Y. Okuzu, T. Kawata, K. Goto, T. Shimizu, Y. Shimizu,T. Kawai, M. Hayashi, K. Morizane, M. Imamura, N.Ikeda, Y. Takaoka and S. Matsuda, J. Biomed. Mater.Res. B, 109, 238–245 (2020).6) N. Ikeda, S. Fujibayashi, S. Yamaguchi, K. Goto, B.Otsuki, T. Kawai, T. Shimizu, Y. Okuzu, K. Masamoto,051015202530h e-max (nm)051015202530E (GPa)AAverage pressure (MPa)050100150200(a) (b) (c)B A B A B250300Fig. 9. Apparent Young’s modulus (a), indentation depth at max. load (b) and average pressure at max. loadwithin the elastic deformation region (c) of samples A and B.0 10 20 30 40 50 60 70 80h (nm)20040060080010001200P (μN)0 10 20 30 40 50 60 70 80h (nm)(3P)/(4R0.5 h1.5 )  (GPa)1020304050(a) (b)he-maxSampleA-1A-2B-1B-2Fig. 8. Typical indentation load P versus penetration depth h curves for samples A and B (a). (b) showsrelationship between (3P)/(4R0.5h1.5) and h for each curve of (a).Hashimoto et al.: Development of segmented titanium oxynitride layer with spinform surfaces exhibiting antibacterial propertiesJCS-Japan688Y. Shimizu, Y. Takaoka and S. Matsuda, Biomat. Adv.,138, 212952 (2022).7) S. Yamaguchi, P. T. M. Le, S. A. Shintani, H. Takadama,M. Ito, S. Ferraris and S. Spriano, Nanomaterials-Basel,11, 2199 (2021).8) M. Kawashita, N. Matsui, T. Miyazaki and H. Kanetaka,Colloid. Surface. B, 111, 503–508 (2013).9) S. Rtimi, C. Pulgarin, M. Bensimon and J. Kiwi, Sol.Energy, 93, 55–62 (2013).10) S. Hawi, S. Goel, V. Kumar, O. Pearce, W. N. Ayre andE. P. Ivanova, ACS Appl. Nano Mater., 5, 1–17 (2022).11) K. A. Whitehead, D. Rogers, J. Colligon, C. Wright andJ. Verran, Colloid. Surface. B, 51, 44–53 (2006).12) N. Ogawa, K. Sato, K. Sunada, H. Ishiguro, H. Kojimaand T. Ito, J. Photopolym. Sci. Tec., 35, 213–217(2022).13) L. Ploux, K. Anselme, A. Dirani, A. Ponche, O. Sopperaand V. Roucoules, Langmuir, 25, 8161–8169 (2009).14) E. Fadeeva, V. K. Truong, M. Stiesch, B. N. Chichkov,R. J. Crawford, J. Wang and E. P. Ivanova, Langmuir,27, 3012–3019 (2011).15) Y. Inoue, M. Uota, T. Torikai, T. Watari, I. Noda, T.Hotokebuchi and M. Yada, J. Biomed. Mater. Res. A,92, 1171–1180 (2009).16) S. Ferraris, A. Venturello, M. Miola, A. Cochis, L.Rimondini and S. Spriano, Appl. Surf. Sci., 311, 279–291 (2014).17) E. P. Ivanova, J. Hasan, H. K. Webb, G. Gervinskas, S.Juodkazis, V. K. Truong, A. H. F. Wu, R. N. Lamb,V. A. Baulin, G. S. Watson, J. A. Watson, D. E.Mainwaring and R. I. J. Crawford, Nat. Commun., 4, 1–7 (2013).18) N. Yamaguchi, K. Wada, K. Kimura and H. Matsubara,J. Ceram. Soc. Jpn., 111, 883–889 (2003).19) T. Yokoi, N. Yamaguchi, M. Tanaka, D. Yokoe, T. Kato,S. Kitaoka and M. Takata, Mater. Lett., 193, 176–178(2017).20) S. Kitaoka, M. Tanaka, N. Kawashima, T. Ito, D. Yokoe,T. Kato, T. Ogawa, N. Yamazaki, N. Hosoya and T.Nakamura, J. Am. Ceram. Soc., 106, 4863–4876(2023).21) T. Akatsu, S. Numata, T. Demura, Y. Shinoda and F.Wakai, Mech. Mater., 83, 66–71 (2015).22) T. Akatsu, S. Numata, T. Demura, Y. Shinoda and F.Wakai, Mech. Mater., 92, 1–7 (2016).23) M. Hashimoto, T. Ogawa, S. Kitaoka, S. Muto, M.Furuya, H. Kanetaka, M. Abe and H. Yamashita, ActaMater., 155, 379–385 (2018).24) NIST X-ray Photoelectron Spectroscopy Database(SRD 20) Ver. 5.0.25) R. Kiribayashi, K. Sunada, Y. Mochizuki, T. Isobe, S.Matsushita, T. Nagai, H. Ishiguro and A. Nakajima,J. Ceram. Soc. Jpn., 131, 117–125 (2023).26) R. Nakane, R. Kiribayashi, K. Sunada, Y. Mochizuki, T.Isobe, S. Matsushita, T. Nagai, H. Ishiguro and A.Nakajima, Ceram. Int., 50, 10797–10805 (2024).27) M. Minoshima, Y. Lu, T. Kimura, R. Nakano, H.Ishiguro, Y. Kubota, K. Hashimoto and K. Sunada,J. Hazard. Mater., 312, 1–7 (2016).28) N. Aničić, M. Vukomanović, T. Koklič and D. Suvorov,Small, 14, 1800205 (2018).29) S. Legmairi, S. Meneceur, G. G. Hasan, L. S. Eddine,H. A. Mohammed, F. Alharthi and J. A. A. Abdullah,Nanotechnology, 34, 445101 (2023).30) Y. Bao, W. Wang, W. Cui and G. Qin, Surf. Coat. Tech.,419, 127296 (2021).31) E. Fadeeva, V. K. Truong, M. Stiesch, B. N. Chichkov,R. J. Crawford, J. Wang and E. P. Ivanova, Langmuir,27, 3012–3019 (2011).32) A. Jemat, M. J. Ghazali, M. Razali and Y. Otsuka,BioMed Res. Int., 2015, 791725 (2015).33) F. H. Rajab, C. M. Liauw, P. S. Benson, L. Li and K. A.Whitehead, Colloid. Surface. B, 160, 688–696 (2017).34) S. Spriano, S. Yamaguchi, F. Baino and S. Ferraris, ActaBiomater., 79, 1–22 (2018).Journal of the Ceramic Society of Japan 132 [12] 681-689 2024 JCS-Japan689