E. Schafer
Department of Operative Dentistry, University of Munster, Munster, Germany.

Introduction.
In recent years, nickel-titanium (NiTi) alloy has been successfully used in the manufacture of endodontic instruments (Schafer 1997, Thompson 2000).This alloy, named Nitinol, consists of approximately 55% nickel and 45% titanium by weight (Schafer 1997, Thompson 2000). Owing to their substantially increased flexibility compared to stainless steel instruments, NiTi files are reported to be particularly suitable for preparing curved root canals (Pettiette et al. 1999, Thompson 2000, Bergmans et al.2001, Peters et al.2001). NiTi instruments were shown to have three times the elastic flexibility in bending and torsion of the stainless steel files (Walia et al.1989, Tepel et al.1997). Nitinol alloys exhibit superelastic behaviour, allowing them to return to their original shape upon unloading following deformation (Schafer1997, Thompson 2000).
Manufacturing of NiTi files is more complex than that of stainless steel instruments as the files have to be machined rather than twisted (Thompson 2000). Owing to its superelasticity, it is impossible to twist a NiTi blank counterclockwise in order to produce a spiral since NiTi alloys undergo almost no permanent deformation (Schafer 1997, Thompson 2000). More likely, they will fracture when being extensively twisted in order to produce a spiral. However, it is known that grinding of nickel-based alloys is difficult because considerable wear of the milling head occurs within a short time (Schafer 1997). This leads to structural defects especially on the cutting edges of NiTi instruments that may compromise the cutting efficiency of these instruments (Schafer 1997, Haikel et al. 1998, Thompson 2000).At the same time, metal flas hand surface heterogeneity might allow the NiTi to corrode (Serene et al. 1995).
Therefore, some surface engineering techniques have been used to improve the surface hardness and the corrosion resistance of NiTi instruments (Lee et al. 1996, Rapisarda et al. 2000; 2001). The surface hardness of Nitinol can be increased by ion implantation of boron (Lee et al.1996). Ionic implantation of nitrogen ions creates a surface layer of titanium nitride (TiN)t hat enhances surface hardness, cutting efficiency and wear resistance (Rapisarda et al. 2000; 2001). Furthermore, most recently it has been shown that a TiN hard coating created by a physical vapour deposition (PVD) cathodic arc evaporation technique increased the cutting efficiency of NiTi files significantly (Schafer in press).To date, no studies have examined the effects of repeated sterilization on the cutting efficiency of these TiN-coated NiTi instruments.
The purpose of the present study was to investigate potential alterations in cutting efficiency when conventional and TiN-coated NiTi K-files that had undergone PVD cathodic arc evaporation coating were exposed to repeated sterilization under autoclave. Moreover, the alterations in cutting of both uncoated and TiN coated NiTi files after exposure to sodium hypochlorite (NaOCl) and repeated autoclave sterilization were also evaluated.

Materials and methods.

Instruments.
A total of 96 new NiTi (size 35) hand K-files (Naviex, Brasseler, Savannah, GA, USA) we re examined. All instruments were taken from a single batch (081501515). The instruments were randomly divided into two groups (groups A and B)o f 48 instruments each. Whilst the instruments of group B were exposed to PVD coating creating a TiN layer with a thickness of 1.5 mm, as previously described (Schafer in press), the files of group A were not coated. Subsequently, the instruments of groups A and B were randomly divided into four subgroups of12 instruments each.

1. Group A.1 (experimental): Uncoated instruments were exposed to five cycles of sterilization under autoclave.
2. Group A.2 (experimental): Uncoated instruments were exposed to 10 cycles of sterilization under autoclave.
3. Group A.3 (experimental): Uncoated instruments were immersed in 5.25% NaOCl for 30 min, rinsed for 10 min in tap water, allowed to dry and thereafter exposed to five cycles of sterilization under autoclave.
4. Group A.C (control): Uncoated instruments were not exposed to the sterilization process.
5. Group B.1 (experimental): TiN-coated instruments were exposed to five cycles of sterilization under autoclave.
6. Group B.2 (experimental): TiN-coated instruments were exposed to 10 cycles of sterilization under autoclave.
7. Group B.3 (experimental): TiN-coated instruments were immersed in 5.25% NaOCl for 30 min, rinsed for 10 min in tap water, allowed to air dry and thereafter exposed to five cycles of sterilization under autoclave.
8. Group B.C (control): TiN-coated instruments were not exposed to the sterilization process.

The autoclave used was a Aesculap Automat 356 (Aesculap, Tuttlingen, Germany). Each cycle of sterilization was performed at 134 8C and 2 bar for 30 min without any corrosion protection. Between the single cycles of sterilization, the instruments were allowed to dry and cool to room temperature.
The TiN coating was created using a PVD technology described previously (Schafer in press).The instruments were fixtured in a vacuum coating chamber and preheated. After heating, an argon ion bombardment cycle commenced in order to clean the surface of the instruments prior to depositing the coating. The substrate temperature was 180 8C. Cathodic arc evaporation was used to create a highly ionized plasma. Following the cleaning process, an arc was struck on multiple titanium cathodes positioned inside. The arc flash evaporated titanium which was attracted to the negatively charged instruments (bias = 60 V). Following the creation of a thin layer of about100 nm of pure titanium, which acted as an adhesive layer, a second layer of TiN was created with a thickness of 1.5 mm. Thereafter, nitrogen gas was introduced at a low partial pressure into the coating chamber which reacted with the titanium to form titanium nitride.

Cutting efficiency.
The spiral of a K-file establishes a cutting angle, i.e. an angle of the flutes to the long axis of the instrument, which is less than 458 (Schafer 1997). Therefore, the cutting efficiency of all instruments was determined in a rotary working motion since these instruments are primarily designed to be used in this way (Schafer 1997).
Cutting efficiency of all instruments was determined by means of a specially designed, computer-driven testing device. The function of this test apparatus has been described in detail in previous studies (Tepel et al. 1995a,b).Special plastic samples with a cylindrical canal having well-defined abrasive properties were used and the maximum penetration depth of the instruments into the lumen was the criterion for cutting efficiency and the basis for the comparison (Tepel et al. 1995a,b). The cylindrical lumen was 17 mm long and the diameter of this lumen was 0.40 mm (Tepel et al.1995a,b) .
The data established for uncoated and coated instruments were analyzed separately. Anova was used to compare maximum penetration depths across the four groups. Thereafter, Bonferroni-adjusted pair wise t-tests were used to evaluate which groups differed from the accompanying control group. The level of significance was set at P < 0.05.

For all groups, the mean maximum penetration depths and standard deviations are given in Table 1. The nonsterilized TiN-coated (group B.C)fifiles reached significantly greater maximum penetration depths than the nonsterilized uncoated instruments (group A.C; P < 0.001).
The sterilized uncoated files (groups A.1-A.3) displayed lower maximum penetration depths than the never-sterilized control instruments (group A.C). Out of 12 NaOCl-treated uncoated instruments (group A.3), two files fractured during the determination of cutting efficiency; no other instrument separations were recorded.
Anova revealed no significant difference amongst the four groups of the TiN-coated instruments (P = 0.110). Independent of the number of sterilization cycles or the immersion in NaOCl prior to sterilization, the maximum penetration depths of the coated instruments (Fig.1) showed no statistically significant difference in comparison with the penetration depths of the control instruments (P > 0.05).

Table 1. Mean maximum penetration depths (mm) and SD achieved by the instruments of the different groups (for each group n=12 instruments).



Figure 1. Notched boxplots of the maximum penetration depths (mm) achieved by the TiN-coated instruments of the different groups.
Group B.1: Five cycles of sterilization; B.2:10 cycles of sterilization; B.3: NaOCl treatment prior to five cycles of sterilization; B.C: Nonsterilized controls.

Discussion.
The PVD technique is a common method to deposit wear resistant thin film coatings on surgical or dental instruments (Grohmann & Mathey 1991, Schafer in press). PVD includes reactive magnetron sputtering, ion plating and arc evaporation (Smith1995). The cathodic arc evaporation technique is often used to create hard coatings including TiN, TiC, TiCN and TiAlN (Smith 1995). Using this technique it is possible to deposit a fine grained TiN film on instruments at comparatively low temperatures (Grohmann &Mathey1991, Schafer in press).Coating thickness ranges from1to 7 mm and it is possible to obtain surface hardness of about 2200 Vickers units (Grohmann & Mathey1991).
The test for determination of cutting efficiency employed in the present study has been described in detail (Schafer 1995, Tepel et al. 1995a,b). Tolerances in the diameter of the files or in the diameters of the specimens’ lumen did not affect the results because the lumen diameter of the cylindrical canal was 0.40 mm whilst the tapered instruments had a tip diameter of 0.35 mm. Thus, the instruments, which did not rotate at this time, could passively penetrate the lumen to a certain extent without contacting the canal wall. As soon as the instrument came into circumferential contact with the wall of the specimens, the motor of the testing device was switched on and the rotating instrument removed material and penetrated actively, i.e. deeper into the lumen until it was blunt and the maximum penetration depth was reached. Since only the passive penetration depth depends on the aforementioned tolerances, this was the reference (zero) line for the determination of the maximum penetration depth and, therefore, for determination of cutting efficiency in a rotary working motion (Tepel et al.1995a,b).Admittedly, the rotary motion generated by the testing device did not reflect a clinical situation in which the endodontic instrument is required to flex whilst it rotates, for example, when enlarging curved root canals. Thus, the results of the present investigation allow no prediction whether or not the TiN coating withstands repeated flexural stress, resulting in fatigue.
Nevertheless, according to the present results, the PVD coating technique increased the cutting efficiency of NiTi files. The TiN-coated instruments that were not sterilized (group B.C)h ad a significant increase (P < 0.001)i n cutting efficiency by up to 25.4% in comparison to the uncoated instruments of the same batch (group A.C)t o corroborate the results of previous studies (Rapisarda et al. 2000, Schafer in press).
Repeated sterilization of endodontic instruments using either steam sterilization or autoclave have been shown to have a minor influence on the cutting efficiency of stainless steel instruments (Schafer 1995, Haikel et al. 1996). Moreover, the bending and torsional properties of either NiTi or stainless steel instruments were not consistently affected by different sterilization procedures (Iverson et al. 1985, Silvaggio & Hicks 1997, Canalda-Sahli et al.1998, Svec & Powers1999).
The sterilized uncoated files (groups A.1 to A.3) had a statistically significant loss of cutting efficiency in comparison with instruments that were not sterilized (control group; P < 0.05). Files that underwent five cycles of sterilization experienced a 16.1% reduction in cutting efficiency compared with the control files (Fig. 2). After 10 cycles of sterilization, the cutting efficiency was reduced further by up to 50.8% (Table 1). The present findings are in good agreement with those of two other studies (Schafer1995, Rapisarda et al.1999).

Figure 2. Notched boxplots of the maximum penetration depths (mm) achieved by the uncoated instruments of the different groups.
Group A.1: Five cycles of sterilization; A.2: 10 cycles of sterilization; A.3: NaOCl treatment prior to five cycles of sterilization; A.C: Nonsterilized controls.

The explanation for the undesirable effect of sterilization might be that sterilized instruments show changes in their outer surfaces and in their chemical composition (Rapisarda et al.1999). It was observed that sterilization produced an increase in titaniumoxide in the near surface layer of the instruments (Rapisarda et al.1999).
In the present investigation, two out of12 NaOCl-treated uncoated instruments separated during the determination of cutting efficiency. Since clinically, size 35 hand instruments rarely really separate, this might be owing to the torsional load in the specific test rather than due to an adverse effect of the sterilization procedures or the NaOCl treatment on the torsional properties of these instruments.
According to the present results, 30-min immersion of uncoated NiTi instruments in NaOCl followed by five cycles of sterilization (group A.3) resulted in a statistically significant loss of cutting efficiency (17.4%; P < 0.05)( Table 1). The results reveal that the maximum penetration depths of the files after immersion in NaOCl followed by repeated sterilization did not differ from those after five cycles of sterilization without NaOCl treatment (3.06 mm vs. 3.11mm; Table 1). Thus, the present results seem to confirm a previous study in that NaOCl treatment without any sterilization cycle seems to have no or only insignificant adverse effect on the cutting efficiency of NiTi instruments (Haikel et al. 1998). These findings are in good agreement with the results of Busslinger et al. (1998) who showed that NiTi was resistant to 1% solution of NaOCl, whilst a statistically significant amount of titanium was lost from the NiTi instruments after immersion times of 30 and 60 min in 5% NaOCl. Just as in the present study, the authors concluded that it is doubtful if this is of any clinical significance, since clinically, files do not have an ‘in situ’ contact time of 30 min with NaOCl (Busslinger et al.1998). Thus, according to these findings and the present results, it can be assumed that only the sterilization cycles displayed an adverse effect on the cutting efficiency of the files.
It has been stated that stainless steel hand files should be considered disposable instruments because they wear rapidly when used on dentine (Kazemi et al. 1995).These authors proposed that from a cost-effective point of view, files should be considered disposable because it is impossible to ensure that files are optimally efficient after use (Kazemi et al. 1995). Certainly, the same is true for uncoated NiTi hand instruments, which are also known to wear rapidly (Walia et al. 1989, Thompson 2000). However, further research is necessary to provide data on how often TiN-coated NiTi instruments can be reused without a considerable decline in efficiency.

References.

Bergmans L, van Cleynenbreugel J,Wevers M, Lambrechts P (2001)   Mechanical root-canal preparation with NiTi rotary instruments: rationale,   performance and safety. Status report for the American Journal of Dentistry.   American Journal of Dentistry14, 324-33.
  Busslinger A, Sener B, Barbakow F (1998) Effects of sodium hypochlorite on   nickel-titanium Lightspeed instruments. International Endodontic Journal   31, 290-4.
  Canalda-Sahli C, Brau-Aguade E, Sentis-Vilalta J (1998) The effect of sterilization   on bending and torsional properties of K-files manufactured with different metallic   alloys. International Endodontic Journal 31, 48-52.
  Grohmann FR, Mathey Y (1991) PVD coating in dentistry. Zahnarztliche   Mitteilungen 81, 30-1.
  Haikel Y, Serfaty R, Lwin T-T, Bleicher P, Allemann C (1996) Effects of cleaning,   disinfection, and sterilization procedures on the cutting efficiency of endodontic   files. Journal of Endodontics 22, 657-61.
  Haikel Y, Serfaty R,Wilson P, Speisser JM, Allemann C (1998) Cutting efficiency   of nickel-titanium endodontic instruments and the effect of sodium hypochlorite   treatment. Journal of Endodontics 24, 736-9.
  Iverson G, von Fraunhofer J, Herrmann J (1985) The effect of various sterilization   methods on the torsional strength of endodontic files. Journal of Endodontics11,   266-8.
  Kazemi RB, Stenman E, SpOngberg LSW (1995) The endodontic file is a disposable   instrument. Journal of Endodontics 21, 451-5.
  Lee DH, Park JB, Saxena A, SereneTP (1996) Enhanced surface hardness by boron   implantation in Nitinol alloy. Journal of Endodontics 22, 43-6.
  Peters OA, Schonenberger K, Laib A (2001) Effects of Ni-Ti preparation techniques   on root-canal geometry assessed by microcomputed tomography. International   Endodontic Journal 34, 221-30.
  Pettiette MT, Metzger Z, Phillips C, Trope M (1999) Endodontic complications   of root-canal therapy performed by dental students with stainless-steel K-files   and nickel-titanium hand files. Journal of Endodontics 25, 230-4.
  Rapisarda E, Bonaccorso A,TripiTR, Condorelli GG (1999) Effect of sterilization   on the cutting efficiency of rotary nickel- titanium endodontic files. Oral   Surgery, Oral Medicine Oral Pathology, Oral Radiology and Endodontics 88, 343-7.
  Rapisarda E, Bonaccorso A, Tripi TR, Condorelli GG, Torrisi L (2001) Wear   of nickel-titanium endodontic instruments evaluated by scanning electron microscopy:   effect of ion implantation. Journal of Endodontics 27, 588-92.
  Rapisarda E, Bonaccorso A, Tripi TR, Fragalk I, Condorelli GG (2000) The   effects of surface treatments of nickel-titanium files on wear and cutting efficiency.   Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics   89, 363-8.
  Schafer E (1995) Effects of various sterilization procedures on the cutting   efficiency of root-canal instruments. Deutsche Zahnarztliche Zeitschrift   50, 150-3.
  Schafer E (1997) Root-canal instruments for manual use: a review. Endodontics   and DentalTraumatology13, 51-64.
  Schafer E (in press) Effect of physical vapor deposition (PVD) on cutting   efficiency of nickel-titanium files. Journal of Endodontics, in press.
  Serene TP, Adams JD, Saxena A (1995) Nickel-Titanium Instruments. Applications   in Endodontics. St. Louis, MO, USA: Ishiyaku EuroAmerica Inc.
  Silvaggio J, Hicks ML (1997) Effect of heat sterilization on the torsional   properties of rotary nickel-titanium endodontic files. Journal of Endodontics   23, 731-4.
  Smith DL (1995) Thin Film Deposition: Principles and Practice. New York,   USA: McGraw-Hill.
  Svec TA, Powers JM (1999) Effects of simulated clinical conditions on nickel-titanium   rotary files. Journal of Endodontics 25, 759-60.
  Tepel J, Schafer E, Hoppe W (1995a) Root canal instruments for manualuse:   cutting efficiency and instrumentation of curved canals. International Endodontic   Journal 28, 68-76.
  Tepel J, Schafer E, Hoppe W (1995b) Properties of endodontic hand instruments   used in rotary motion. Part1. Cutting efficiency. Journal of Endodontics   21, 418-21.
  Tepel J, Schafer E, HoppeW (1997) Properties of endodontic hand instruments   used in rotary motion. Part 3. Resistance to bending and fracture. Journal   of Endodontics 23, 141-5.
  Thompson SA (2000) An over view of nickel-titanium alloys used in dentistry.   International Endodontic Journal33, 297-300.
  Walia H, Costas J, Brantley W, Gerstein H (1989) Torsional ductility and   cutting efficiency of the Nitinol file. Journal of Endodontics15, 174(abstract).