DENTAL-2444; No. of Pages 11

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

Available online at

ScienceDirect journal homepage:

A new testing protocol for zirconia dental implants Clarisse Sanon a,b , Jérôme Chevalier a,c,∗ , Thierry Douillard a , Maria Cattani-Lorente d , Susanne S. Scherrer d , Laurent Gremillard a a

INSA-Lyon, UMR CNRS 5510 (MATEIS), 7 avenue Jean Capelle, 69621 Villeurbanne Cedex, France University Lyon 1, Faculty of Odontology, Department of Biomaterials, 11, Rue Guillaume Paradin, 69372 Lyon Cedex 08, France c Institut Universitaire de France, 103 bd Saint-Michel, Paris 75005, France d University of Geneva, School of Dental Medicine, Department of Prosthodontics-Biomaterials, 19 rue Barthélémy-Menn, CH 1205 Geneva, Switzerland b

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objectives. Based on the current lack of standards concerning zirconia dental implants, we

Available online xxx

aim at developing a protocol to validate their functionality and safety prior their clinical


specific behavior of zirconia in terms of phase transformation.

use. The protocol is designed to account for the specific brittle nature of ceramics and the Zirconia

Methods. Several types of zirconia dental implants with different surface textures (porous,


alveolar, rough) were assessed. The implants were first characterized in their as-received

Phase transformation

state by Scanning Electron Microscopy (SEM), Focused Ion Beam (FIB), X-Ray Diffraction


(XRD). Fracture tests following a method adapted from ISO 14801 were conducted to evalu-


ate their initial mechanical properties. Accelerated aging was performed on the implants,


and XRD monoclinic content measured directly at their surface instead of using polished samples as in ISO 13356. The implants were then characterized again after aging. Results. Implants with an alveolar surface presented large defects. The protocol shows that such defects compromise the long-term mechanical properties. Implants with a porous surface exhibited sufficient strength but a significant sensitivity to aging. Even if associated to micro cracking clearly observed by FIB, aging did not decrease mechanical strength of the implants. Significance. As each dental implant company has its own process, all zirconia implants may behave differently, even if the starting powder is the same. Especially, surface modifications have a large influence on strength and aging resistance, which is not taken into account by the current standards. Protocols adapted from this work could be useful. © 2014 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.

Corresponding author at: INSA-Lyon, UMR CNRS 5510 (MATEIS), 7 avenue Jean Capelle, 69621 Villeurbanne Cedex, France. E-mail address: [email protected] (J. Chevalier). 0109-5641/© 2014 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11


ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx



Oral implants offer an effective treatment for replacement of missing teeth. Since the pioneering works of Brånemark in the 60’s [1], several millions of titanium implants have been produced. It is reported that the oral implant number will grow at a rate of 6% per year from 2010 to 2015, since there is a tendency to propose more and more implants to patients to improve their quality of life (aesthetic, but also mastication and long-term stability) [2]. Current long-term clinical investigations (more than 10 years of follow-up) report very favorable survival rates, which places titanium and its biomedical alloys as the gold standard [3]. However, in some cases, the greyish color of a titanium implant may be perceived through the peri-implant mucosa causing some aesthetic drawbacks [4]. Furthermore, in rare cases, metals (including titanium) may induce sensitization or allergic reactions [5,6]. Finally, some patients also ask for completely metal-free dental reconstructions. Thus, implants fabricated with ceramic materials are gaining popularity and might have a certain clinical and industrial success if they prove to be strong enough, stable over time and well integrated in the jawbone. Especially, yttriadoped zirconia ceramics (often referred as 3Y-TZP, standing for 3 mol.% Yttria doped Zirconia Tetragonal Polycristals) are often presented as the alternative to titanium [7]. These ceramics possess good mechanical strength [8], excellent tissue compatibility and show osseointegration comparable to that of titanium [9]. A further advantage of zirconia is the reduced formation of plaque [10]. Moreover, the white-opaque ZrO2 ceramic better resembles the tooth in terms of color, and thus provides good esthetics even with a thin gingiva or with soft-tissue recessions. Worldwide, there are more than 10 companies producing zirconia dental implants and each manufacturer develops its process, its implant design and its own surface features to promote osseointegration. It is generally accepted that rough surfaces improve osseointegration and favor mechanical anchorage with bone. Several strategies are explored to process rough or porous surface implant i.e. machining, acid etching, sandblasting, molding or coating with a porous layer [11–13]. Although zirconia has good initial mechanical properties (high fracture toughness and bending strength), it remains a ceramic material with a significant sensitivity to surface defects. The above-mentioned surface treatments may generate cracks and/or defects which could be detrimental for mechanical properties of these zirconia implants. Moreover, screw design allows mechanical anchorage of dental implant into the bone but is challenging for ceramics material because of stress concentration at sharp edges [14]. All these aspects are poorly documented in the recent literature on zirconia dental implants. It has also to be recalled at this stage that 3YTZP was introduced as an implant biomaterial (femoral heads) in orthopedics over 30 years ago, but was abandoned after 15 years of use, after a series of failures in specific batches manufactured with a new process [15]. Zirconia is a complex material because it is meta-stable at room temperature. On the one hand, its excellent mechanical properties (the best of oxide ceramics) are due to the transformation of metastable tetragonal grains to the monoclinic phase under stress (for

example in the vicinity of a crack). The development of this transformation zone is accompanied by an increase of crack resistance, which is known as phase transformation toughening [16]. On the other hand, this meta-stability leads to a possible transformation of grains in contact with water (or body fluids) with time. This phenomenon is often referred as to Low Temperature Degradation (LTD) or aging. Aging is a progressive tetragonal to monoclinic transformation at the surface triggered by the presence of water [16], which often results in surface roughening and micro cracking and thereby potentially decreases the device physicochemical and mechanical properties. The experience of zirconia in orthopedics field gave some important indications on how aging may proceed, on the potential impact of the transformation and on the conditions by which it may be triggered. The transformation proceeds from the surface in contact with water to the bulk of the material. The kinetics by which the transformation occurs is highly dependent on process conditions and resulting microstructure [17]. Surface modifications for example may have a positive effect on bone apposition and bone in-growth, but also could facilitate the water penetration into the bulk and/or lead to a modification of the stability of the tetragonal phase under humid atmosphere. Except few recent works, including one from the authors of the current paper [18], the risk of lifetime reduction associated to surface modification of implants is barely discussed. There is today no standardized protocol that allows assessing the mechanical properties of the implant, to determine the aging kinetics and the effects of aging on the mechanical properties for a given type of implant. The only ISO standard [19] concerning medical grade zirconia is based on mechanical strength and aging kinetics measured on bending bars or discs, which are polished and therefore not relevant for dental implants. To bridge this gap, we aim at proposing a protocol to validate the functionality and safety of zirconia implants prior their clinical use. The protocol is designed to account for the specific brittle nature of ceramics (sensitive to surface defects and slow crack growth) and the specific behavior of zirconia in terms of phase transformation.


Materials and method


Implant description

For this research, Axis Biodental provided two types of 3YTZP dental implants processed by injection molding, with either a structured rough surface, which will be referred as ‘Axis-rough’ or with an additional proprietary porous zirconia coating here referred as ‘Axis-alveolar’ in relation with their surface texture. Axis-rough surface was obtained after surface treatment of the mold inner and the alveolar one, after deposition and sintering of a mixture of zirconia powder and polymer beads (patent application EP 1924300 B1). Only the ‘Axis-rough’ implants were commercial implants, while ‘Axis-alveolar’ were prototypes in the development phase. Nobel Biocare provided zirconia implant prototypes with a porous surface (ZiUnite®). The porous surface was achieved after sintering, by coating the endosseous part of the implants with a slurry containing zirconia powder and a pore former

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),


DENTAL-2444; No. of Pages 11

d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

(patent application SE03022539-2). Further sintering of the implants yielded to the burn off of the pore former and to a porous surface. The presence of the coating gave rise to a rough and porous, 15 ␮m-thick surface layer according to the manufacturer. It is to note that all implants were processed from a biomedical grade 3% mol. yttria-stabilized zirconia powder (Tosoh TZ-3Y-E, Tokyo, Japan).


Microstructural characterization

Microstructural aspect of the two types of implants was investigated using a Scanning Electron Microscope (Supra 40, Carl Zeiss AG, Oberkochen, Germany) to analyze the surface and a dual beam Focus Ion Beam (FIB) for further investigation in-depth. FIB acquisitions were made on the endosseous part of the implants. FIB/SEM imaging was performed using a FIB/SEM workstation (NVision 40; Carl Zeiss Microscopy GmbH, Oberkochen, Germany) combining a SIINT zeta FIB column (Seiko Instruments Inc. NanoTechnology, Japan) with a Gemini column. In brief, the FIB uses a liquid metal ion source of Ga+ ions accelerated between 2 and 30 keV that are focused to the surface to cut slices of materials. SEM images are taken simultaneously with the electron beam. FIB/SEM therefore produces two dimensional image datasets that can be used as cross-sections, but that are also suitable for the reconstruction of microstructures in three dimensions. Three-dimensional analysis using FIB tomography is essentially a two-step process. After acquisition of the raw data as described above, this dataset is taken offline for further processing and 3D visualization. FIB leads to the production of a stack of assumed equidistant cross sections (distance between two cross-sections: 10 nm) through the analyzed volume. The subsequent image processing workflow can link slices fine alignment, data cropping, image filtering, segmentation/threshold operations, morphological operations, labeling, quantification and visualization. Image processing operations were carried out using the software Fiji (, developed at the National Institutes of Health (Bethesda, USA).


where Xm is the integrated intensity ratio, Im(h k l) is the area of the (h k l) peak of the monoclinic phase and It(h k l) is the area of the (h k l) peak of the tetragonal phase. The experimental volume content of monoclinic phase f was then determined with: f =

1.311 × Xm 1 + 0.311 × Xm


The procedure was generally conducted directly at the surface of the implants, on the threaded (endosseous) area. However, in order to assess the influence of surface preparation on aging kinetics, the kinetics was also measured on sectioned and mirror-polished implants, as it is still recommended by the ISO 13356 standard. Effects of aging were specially examined after 5 h of artificial aging at 134 ◦ C, because this aging duration represents the range of the lifetime expected for endosseous implants. 5 h corresponds to 10–20 years in vivo.


Mechanical characterization

Only Axis implants were characterized in terms of load to failure. Indeed, the objective was not to compare the mechanical strength of different types of implants (they do not present the same design) but more to highlight the potential impact of surface modification on a given type of implant. The mechanical tests were carried out using a procedure based on the ISO 14801 [21] as it simulates the functional loading of an endosseous dental implant body and its prosthetic components under the worst possible in vivo conditions: - The implants were embedded in an epoxy resin at 30◦ angulation with respect to the vertical axis. The ISO 14801 recommends a material with a Young’s modulus higher than 3 GPa. Preliminary results led us to choose a resin with a higher stiffness of 11 GPa (RenCast CW 20/HY 49) to avoid any viscoelastic deformation during loading. - The implants were embedded up to a distance of 3 mm below the nominal level specified by the implant manufacturers, to simulate bone resorption.

Aging kinetics

Aging tests were performed in water steam at 134 ◦ C, under 2 bars pressure for durations up to 100 h. It has been reported from our previous records that one hour at 134 ◦ C would roughly correspond to 2 years at 37 ◦ C. This is a rough estimation that can be debated but which gives an idea of treatment durations relevant for the application. Monoclinic content was measured at the surface the endosseous part of the implants by an X-ray diffraction (XRD) technique (CuK␣ radiation) in a −2 mode (2 ∈ [27◦ −33◦ ]) on a Brüker D8 Advance (Brüker, Karlsruhe, Germany) instrument (scan speed of 0.2◦ /min and a step size of 0.02◦ ). Monoclinic content was then calculated by using the formalism given by Garvie and Nicholson:

Xm =


Im(1¯ Im(1¯

1 1)

+ Im(1

1 1) + Im(1

1 1)

1 1)

+ It(1

(1) 0 1)

The implants were loaded to failure at a crosshead speed of 1 mm/min to limit Slow Crack Growth during loading. 15 ‘Axis-Rough’ and 15 ‘Axis-Alveolar’ implants were tested, either in the as-received state (5 implants of each), or after 5 h or 100 h of accelerated aging at 134 ◦ C in water steam.


Fractographic analysis

Failed specimens underwent fractographic inspection. Failure surfaces were first observed with a stereomicroscope (Olympus XZ9) for the overall orientation of the crack direction and propagation visualized by the presence of a compression curl, larger and finer hackle all pointing back to the area of crack origin. Detailed crack features were viewed with a scanning electron microscope (SEM, FEI, XL30 FEG, SUPRA, Eindhoven, The Netherlands).

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11


ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx



3.1. Microstructural features of the as-received implants SEM pictures of the three types of implants and their typical surface features are given in Fig. 1, for two magnifications. ‘Axis-Rough’ and ‘Axis-Alveolar’ implants exhibit the same thread design, which differs from ZiUnite® implants. ‘AxisAlveolar’ surface is characterized by craters of less than 100 microns, distributed in the valleys of the threads. Their dimensions are related to the size of the pore formers (polymer beads) used during the process. The main information from low magnification image is the detection of surface cracks already existing on the as-received ‘Axis-Alveolar’ implant prototypes, in the valleys of the threads. The two other types of implants do not show such cracks. High magnification pictures show the microstructural features of the three surfaces. ZiUnite® implants are characterized by a micro-porous surface, with pores ranging between 2 and 5 microns. Ra was of 1.2 ± 0.3 ␮m as measured in a previous work. ‘Axis-Rough’ implants showed no surface porosity and a rounded relief with a Ra = 1.5 ± 0.3 ␮m, as a negative of the injection mold. Fig. 1f gives an example of the craters at the surface of Axis-Alveolar implants. These implants are the roughest (Ra = 11 ± 2 ␮m), obviously. FIB low magnification pictures of Fig. 2 indicate the location of the ionic sectioning for each type of implant. ZiUnite® implants exhibit a porous layer of about 10–15 ␮m on a dense body. The porosity of this layer is more visible at high magnification (Fig. 2b). At still higher magnification (Fig. 3), it is obvious that the majority of the zirconia grains appear unfacetted, with straight boundaries indicating they remained in the tetragonal symmetry after their fabrication. However, grains around pores show clear evidence of twinning, which is a sign of a transformation from tetragonal to the monoclinic phase. In other words, as received ZiUnite® implant prototypes are already partially transformed, preferentially around pores. An example of 3D reconstruction is given in Fig. 3b for this particular type of implants, after isolation of only the surface layer and after contrast segmentation, from an initial volume of 40 ␮m × 30 ␮m × 15 ␮m. Only 3D analysis allows obtaining the porosity content and its interconnectivity. For ZiUnite® implants, the porosity inside the ZiUnite® porous coating is estimated to be 74%, with an interconnection ratio of 0.99 (all pores are interconnected). FIB trenches performed on ‘Axis-Rough’ and ‘Axis-Alveolar’ implants are shown in Fig. 2c to f. Zirconia grains appear as non-transformed for both types of AXIS implants, except very few surface grains that might be already transformed.


Aging kinetics

The aging kinetics of ZiUnite® implants, both measured directly on the porous surface and on a cross-sectioned and polished as recommended by the ISO 13356 are given in Fig. 4a. The aging kinetics are completely different, which a much higher transformation rate when the porous surface is concerned. The surface monoclinic fraction after 5 h of aging reaches more than 60% when measured directly on the porous

Table 1 – Load to failure of the AXIS experimental implants at different aging durations (aging at 134 ◦ C, 2 bars). Aging duration (h)

AXIS-Rough implants

AXIS-Alveolar implants


Average: 596 N St. Dev.: 74 N

Average: 392 N St. Dev.: 27 N


Average: 568 N St. Dev.: 14 N

Average: 431 N St. Dev.: 59 N


Average: 730 N St. Dev.: 69 N

Average: 417 N St. Dev.: 41 N

surface of the implants, while it reaches only 22% when measured on a cross-sectioned and polished surface inside the implant. In other words, the protocol recommended by ISO 13356 does not account for the real transformation rate of the surface of the implants, which can be highly different than that of the bulk. This will be discussed later, but it led us to continue with measurements only on the as received surfaces. Fig. 4b shows the evolution of the surface monoclinic content (i.e. directly on the threaded surface) for the three types of implants. Even though likely processed with the same powder, there is a large difference between Axis and ZiUnite® experimental implants. No significant difference is observed between the two types of Axis implants.


Microstructural features of aged implants

Fig. 5 shows FIB cross-sections taken near the surface of each type of implants, after an aging duration of 5 h (roughly corresponding to 10 years at 37 ◦ C). The ZiUnite® implant appears the most transformed at the surface, in agreement with XRD data. It appears that all the porous coating is transformed, up to a depth of 10–15 microns, which is again consistent with the fact that XRD monoclinic content reaches its plateau at that duration, because the penetration depth of the X-Ray in a porous zirconia with a relative density of 26% is estimated to be of 15 microns (calculations with AbsorbDX software, DIFFRACplus BASIC Evaluation Package, Brüker, Karlsruhe, Germany). In contrast, AXIS implants only show limited transformation with a depth of less than 1 micron.


Mechanical characterization

Table 1 summarizes the load to failure (average value and standard deviation) of the AXIS experimental implants, in the as-received state or after 5 h or 100 h of accelerated aging at 134 ◦ C. The most significant result is that the average value of the load to failure of the AXIS-rough implants is about 50% higher than that of the AXIS-alveolar experimental ones. Given the low number of implants tested (only five per modality), it is difficult to conclude on the influence of aging on the load to failure of the implants, but a tendency towards a slight increase with aging time is observed. Aging, performed on the above-mentioned conditions and for these specific implants, therefore does not lead to strength degradation. It has to be stated at this point that this slight increase in strength with aging time is specific of the implants tested in this work and must not be taken as a general rule in zirconia, as aging

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx


Fig. 1 – SEM pictures of the three types of implants and their surface features.

can be associated to an increase or a decrease in mechanical properties, depending on microstructure and surface treatments [22,23].


(Fig. 6c,d), a mirror zone was sometimes recognizable at the crack origin. Overall, the crack starter zone was larger for the Axis-alveolar (approximately 100–200 microns) than for the Axis-rough specimens (less than 50 microns).

Fractographic analysis

4. Fig. 6 shows typical fractographic images obtained for the two types of Axis experimental implants, after load to failure tests. The crack origin was located at the outer rim on the convex side between two threads and easy to recognize due to the presence of larger hackle nearby the starter crack for the Axis-alveolar specimens (Fig. 6a,b). On the Axis-rough


4.1. On the relevance of current ISO standard to qualify aging of dental implants? The first lesson of the results presented above is the unsuitability of the current ISO standards to evaluate the finished

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11


ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

Fig. 2 – FIB/SEM pictures of the surface of as-received implants.

products versus their sensitivity to aging. Indeed, aging kinetics measured on polished bulk samples are to a very large extent different from those measured at the surface of the implants with a porous surface. This does not mean that ISO 13356 is not straightforward, but it limits it use to qualify the intrinsic aging sensitivity of the material rather than the one of the product, which exhibits generally a different surface than that required by the standard. The present set of results show that the porous ZiUnite® surface degrades the aging sensitivity of the product, when compared to a polished, dense surface. Other examples are available in the literature for which a surface modification (e.g. machining or rough polishing) may improve its resistance to aging. The simple

and obvious protocol used in this work (e.g. aging kinetics measured directly at the surface of the implants) should therefore be added to existing standards to avoid any underestimation or overestimation of their real aging rate.


Effect of surface modification on aging kinetics

As already demonstrated in previous publications, we show that different surface features lead to highly different aging kinetics, in particular at the initial stage of the t-m transformation [17,24]. Dual-Beam FIB/SEM appears once again as a powerful tool to highlight microstructural changes and investigate how the transformation initiates and propagates inside

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx


Fig. 3 – (a) FIB/SEM picture of a ZiUnite implant (as-received) at higher magnification, showing transformation of grains around pores even before accelerated aging tests. (b) 3D reconstruction of the surface coating the after relevant segmentation, to highlight the porosity features of the material (22 ␮m × 4 ␮m × 12.5 ␮m). Interconnected porosity is in blue whereas isolated pores are in green. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) the material. The main origin of the apparent sensitivity of ZiUnite® prototypes to aging is clearly related to the features of the porous zirconia layer. The interconnected porosity offers a path for the transformation to start at every surface

accessible by water, so that the overall porous layer can be transformed in a short period of time (even partially in the as received state). This is not captured by XRD, which gives only an integrated average monoclinic content over the penetration

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11


ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

Fig. 4 – (a) Comparison of aging kinetics measured directly at the surface of a ZiUnite implant and on a sectioned and polished implant (as recommended by ISO 13356). (b) Aging kinetics of ZiUnite, Axis-Rough and Axis-Alveolar implants, all measured at their surface.

depth of the X-Rays (which depends itself to the density/porosity of the surface), and which makes comparisons impossible on the mode of transformation. XRD conducted on flat surfaces might be useful to characterize and qualify a given zirconia against its ‘intrinsic’ stability versus aging, but not to determine how the transformation may proceed for a given type of implant. XRD gives also a convoluted, average value of the monoclinic content over the penetration depth

of the X-Rays (which depends on the porosity of the material, by the way). Without sophisticated techniques (grazing angles) and mathematical de-convolution [25], XRD hardly gives a full picture of the transformation features, while FIB cross-sections give a clear insight on where the transformation starts and how it proceeds. It is recognized that FIB is still not widely spread, and that, even if more time-consuming and more tricky in terms of preparation, very careful

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx


Fig. 5 – FIB/SEM pictures of the surface of the implants after 5 h of aging at 134 ◦ C. (a) and (b) for ZiUnite showing a entire transformation of the porous coating, (c) and (d) for AXIS ‘Rough’ and ‘Alveolar’ showing only limited transformation. Dashed lines represent the limit of the transformation zone.

cross-sectioning and SEM imaging might be able to give useful insight on how the transformation proceeds in zirconia dental implants.

ratio of strength between the two types of implants is given by: f 1 f 2

4.3. Effect of surface modification on mechanical properties

√  × ac

ac2 ac1


The Axis implants design being of the same, the ratio can be translated into load to failure:

The set of results present in this article confirms that some surface modifications may compromise the load to failure of ceramic products, when they lead to an increase of the critical defect size. For the same geometry of implants, the ‘Axis-Alveolar’ prototypes are fractured for a much lower load than the ‘Axis-Rough’ implants, and this is consistent with a larger initial defect size. Starting with the well-known critical stress intensity factor– defect size relation:

KIC = f ×



where KIC is the critical stress intensity factor (or toughness),  f the strength and ac the critical defect size, and assuming a similar toughness for both type of Axis implants, the

Pf 1 Pf 2


ac2 ac1


In the present experimental case, Pf1 /Pf2 ∼ 1.5 and √ ac2 /ac1 ∼1.4 − 2 (considering the defect sizes estimated with the fractographic analysis), which is highly consistent. This simple mechanical test protocol is therefore able to capture the effect of the initial defect size on implants integrity.

4.4. Evaluation of the fatigue limit from simple load to failure tests Zirconia, as all ceramic materials, is sensitive to Slow Crack Growth. In other words, defects can propagate at low rate even if they are submitted to a stress intensity factor below KIC and lead to delayed failure of the implants. Going into details into

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11


ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

Fig. 6 – Fractography of the two types of AXIS experimental implants, after load to failure tests.

the mechanisms of SCG is beyond the scope of the present paper, and the reader could refer to existing references, such as [26,27]. In brief, SCG occurs in the presence of water or body fluid, and may be accelerated by cyclic fatigue. Fortunately, the presence of a threshold stress intensity factor KI0 , below which no propagation occurs, has been claimed in several inorganic materials, both by atomic scale modeling and experiments [28,29]. The value of KIC and KI0 of 3Y-TZP (with a grain size sim√ ilar to the materials tested in this work) are KIC = 6.0 MPa m √ and KI0 = 3.0 MPa m under cyclic fatigue, respectively [29]. The fatigue limit P0 of an implant could therefore be roughly anticipated from the load to failure Pf by:

of implant design associated to the alveolar structure. With a larger number of experimental implants, statistical evaluation (probability of failure) of the delayed failure risk under a given applied load (i.e. under a given clinical situation) could be anticipated by the mechanical tests proposed in this work. It has to be recalled that such tests are not included in current ISO standards for zirconia as material for dental implants. Such tests are easy to run, with simple testing machines, and would give a good first estimation of the fatigue limit of newly developed zirconia dental implants. As far as lifetime under cyclic conditions is concerned, they could be complemented with real cyclic fatigue tests under the same testing geometry.

0 KI0 P0 = = ∼0.5 Pf f KIC



where P0 and  0 refer to the fatigue load and stress limit, respectively, and Pf and  f are the load and stress to failure during fast fracture tests. In the case of the ‘Axis-Alveolar’ prototypes, the average value of the load to failure is of 396 N, which would mean that roughly 50% of the implants would resist to 198 N (and thus 50% would fail), this load value being on the order of the reported average mastication forces [30]. Given the limit of the present analysis (especially because there is a variability on the implant position and on the mastication forces), it would anyway certainly show that the risk of failure is too important with this combination of design and surface features. The producer, based on the analysis above, has abandoned this type


Zirconia being sensitive to slow crack growth and aging, it is important to include relevant tests in future protocols to insure long-term performance and safety of dental implants. The present work shows the limit of current ISO 13356 standard to insure the reliability and lifetime of zirconia dental implants and the need to include complementary methods to XRD to follow more in details aging kinetics. New protocols should therefore include:

- Aging kinetics performed directly on the implant surfaces of the implants, and not on polished samples as it is generally performed in current standards,

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

DENTAL-2444; No. of Pages 11

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 4 ) xxx–xxx

- Direct observation of the transformation zone, in addition to the widespread XRD, such as the use of FIB, or careful cross-sectioning and visualization, - Load to failure under realistic conditions in order to predict the risk of failure for a given clinic situation. For this purpose, the use of a testing geometry such as the one generally used for metallic implants in ISO 14801 would be preferred. This study also shows that zirconia reliability and lifetime can be highly dependent on surface preparation and that every new surface modification should be tested against aging/fracture before clinical use. By the way, it also shows that aging is not systematically associated to a decrease of failure load, but other negative consequences can be speculated (micro-cracking, delamination of the surface and loss of integration in the bone).






Acknowledgements Thanks are due to the CLYM (Centre Lyonnais de Microscopie: for the access to the microscope FIB ZEISS, NVISION 40. CLYM is supported by the CNRS, the “Grand Lyon” and the “Rhône-Alpes Region”. The authors are grateful to Nobel Biocare and Axis Biodental for providing the implants and implant prototypes. The Authors declare no conflict of interest.


[16] [17]


references [19] [1] Branemark PI. Osseointegration and its experimental background. J Prosth Dentistry 1983;50(3):399–410. [2] Babbush CA, Hahn JA, Krauser JT, Rosenlicht JL. Dental implants: the art and science. Maryland Heights: Saunders/Elsevier; 2011. [3] Karoussis IK, Salvi GE, Heitz-Mayfield LJA, Brägger U, Hämmerle CHF, Lang NP. Long-term implant prognosis in patients with and without a history of chronic periodontitis: a 10-year prospective cohort study of the ITI® Dental Implant System. Clin Oral Implants Res 2003;14(3):329–39. [4] Dunn DB. The use of a zirconia custom implant-supported fixed partial denture prosthesis to treat implant failure in the anterior maxilla: a clinical report. J Prosth Dentistry 2008;100:415–21. [5] Vamanu CI, Høl PJ, Allouni ZE, Elsayed S, Gjerdet NR. Formation of potential titanium antigens based on protein binding to titanium dioxide nanoparticles. Int J Nanomed 2008;3(1):69. [6] Sicilia A, Cuesta S, Coma G, Arregui I, Guisasola C, Ruiz E, et al. Titanium allergy in dental implant patients: a clinical study on 1500 consecutive patients. Clin Oral Implants Res 2008;19(8):823–35. [7] Oliva J, Oliva X, Oliva JD. Five-year success rate of 831 consecutively placed Zirconia dental implants in humans: a comparison of three different rough surfaces. Int J Oral Maxillofacial Implants 2009;25(2):336–44. [8] Manicone PF, Rossi Iommetti P, Raffaelli L. An overview of zirconia ceramics: basic properties and clinical applications. J Dentistry 2007;35(11):819–26. [9] Gahlert M, Gudehus T, Eichhorn S, Steinhauser E, Kniha H, Erhardt W. Biomechanical and histomorphometric

[21] [22]




[26] [27]





comparison between zirconia implants with varying surface textures and a titanium implant in the maxilla of miniature pigs. Clin Oral Implants Res 2007;18(5):662–8. Scarano A, Piattelli M, Caputi S, Favero GA, Piattelli A. Bacterial adhesion on commercially pure titanium and zirconium oxide disks: an in vivo human study. J Periodontol 2004;75(2):292–6. Sennerby L, Dasmah A, Larsson B, Iverhed M. Bone tissue responses to surface-modified zirconia implants: a histomorphometric and removal torque study in the rabbit. Clin Implant Dentistry Related Res 2005 2005;7:13–20. Depprich R, Zipprich H, Ommerborn M, Naujoks C, Wiesmann HP, Kiattavorncharoen S, et al. Osseointegration of zirconia implants compared with titanium: an in vivo study. Head Face Med 2008;4, art 30. Yamashita D, Machigashira M, Miyamoto M, Takeuchi H, Noguchi K, Izumi Y, et al. Effect of surface roughness on initial responses of osteoblast-like cells on two types of zirconia. Dental Mater J 2009;28:461–70. Ullman DG. The mechanical design process, Vol. 2. New York:: McGraw-Hill; 1992. Chevalier J, Gremillard L, Virkar A, Clarke DR. The tetragonal-monoclinic transformation in zirconia: lessons learned and future trends. J Am Ceram Soc 2009;92:1901–20. Garvie RC, Hannink RH, Pascoe RT. Ceramic steel? 1975:703–4. Chevalier J, Loh J, Gremillard L, Meille S, Adolfson E. Low-temperature degradation in zirconia with a porous surface. Acta Biomater 2011;7(7):2986–93. Sanon C, Chevalier J, Douillard T, Kohal RJ, Coelho PG, Hjerppe J, et al. Low temperature degradation and reliability of one-piece ceramic oral implants with a porous surface. Dent Mater 2013;29(4):389–97. ISO 13356:2008. Implants for surgery—Ceramic materials based on yttria-stabilized tetragonal zirconia (Y-TZP). ISO 14801:2007. Dentistry-Implants-Dynamic fatigue test for endosseous dental implants. Kosmacˇ T, Oblak C, Marion L. The effects of dental grinding and sandblasting on ageing and fatigue behavior of dental zirconia (Y-TZP) ceramic. J Eur Ceram Soc 2008;28(5): 1085–90. Marro FG, Mestra A, Anglada M. Weibull strength statistics of hydrothermally aged 3 mol% yttria-stabilised tetragonal zirconia. Ceram Int 2014;40(8 Part B):12777–82. Deville S, Chevalier J, Gremillard L. Influence of surface finish and residual stresses on the ageing sensitivity of biomedical grade zirconia. Biomaterials 2006;27(10):2186–92. Gremillard L, Grandjean S, Chevalier J. A new method to measure monoclinic depth profde in zirconia-based ceramics from X-ray diffraction data. J Mater Res 2010;101(1):88–94. Lawn BR. Fracture of brittle solids. 2nd ed. Cambridge: Cambridge University Press; 1993. Olagnon C, Chevalier J, Pauchard V. Global description of crack propagation in ceramics. J Eur Ceram Soc 2006;26:2059–3051. Zhu T, Li J, Lin X, Yip S. Stress-dependent molecular pathways of silica–water reaction. J Mech Phys Solids 2005;53:1597–623. Chevalier J, Aza AHD, Fantozzi G, Schehl M, Torrecillas R. Extending the lifetime of ceramic orthopaedic implants. Adv Mater 2000;12:1619–21. Andreiotelli M, Wenz HJ, Kohal RJ. Are ceramic implants a viable alternative to titanium implants? A systematic literature review. Clin Oral Implants Res, 2010 2009;20(s4):32–47.

Please cite this article in press as: Sanon C, et al. A new testing protocol for zirconia dental implants. Dent Mater (2014),

A new testing protocol for zirconia dental implants.

Based on the current lack of standards concerning zirconia dental implants, we aim at developing a protocol to validate their functionality and safety...
6MB Sizes 3 Downloads 7 Views