Improved electrochemical performance of nitrocarburised stainless steel by hydrogenated amorphous carbon thin films for bone tissue engineering

Abstract

In this study, hydrogenated amorphous carbon thin films, structurally similar to diamond‐like carbon (DLC), were deposited on the surface of untreated and plasma nitrocarburised (Nitrocarburizing‐treated) stainless steel medical implants using a plasma‐enhanced chemical vapour deposition method. The deposited DLC thin films on the nitrocarburising‐treated implants (CN+DLC) exhibited an appropriate adhesion to the substrates. The results clearly indicated that the applied DLC thin films showed excellent pitting and corrosion resistance with no considerable damage on the surface in comparison with the other samples. The CN+DLC thin films could be considered as an efficient approach for improving the biocompatibility and chemical inertness of metallic implants.

1 Introduction

During the last decade, many metallic substrates have been used for biomedical applications such as orthopaedic and dental implants, screws hip nails etc. As a potential candidate, American Iron and Steel Institute 316L stainless steel (SS) has been extensively used in biomedical applications and there are growing research on the enhancement of its functionality in the body [1]. However, some studies have previously shown that SS‐based medical implants may encounter in vivo pitting corrosion [2]. One of the ways for improving corrosion resistance of SS alloys is surface modification strategies. It has been reported that low‐temperature plasma nitriding or nitrocarburising could improve the pitting corrosion of 316L SS if chromium (Cr) nitrides/carbides is not formed on the surface of the implant [3, 4]. The formation of Chromium nitrides/carbides can decrease the corrosion resistance of SS, as a consequence of reduction in Cr content in the matrix and preventing the formation of a continuous passive Cr₂ O₃ layer [5, 6]. Therefore, plasma nitriding or nitrocarburising may not be a suitable method alone for the surface modification of biomedical implants because of probable formation of Cr nitrides/carbides.

Recently, researchers have paid more attention to biocompatible and corrosion resistant thin film coatings for implants [7]. Diamond‐like carbon (DLC) thin films on medical grade metallic substrates could be applied in surgical equipment and implants, since they require high corrosion resistance and biocompatibility [8]. In recent years, biomedical applications of DLC thin films in forms of either hydrogenated amorphous carbon or non‐hydrogenated amorphous carbon coatings have been widely studied due to their unique properties such as high hardness, low friction coefficient, high wear resistance and chemical inertness [9]. This class of thin films has amorphous structure that contains a mixture of sp³, sp², sp¹ bonds and in some cases hydrogen. The physical properties of these thin film coatings are variable by changing the ratio of sp³ (diamond‐like) to sp² (graphite‐like) bonds, and hydrogen content. The challenge which limits the maximum thickness of these coatings is high residual stress leading to delaminate the coating from the substrate. It has been previously reported that a reasonable adhesion of the films to the substrates could significantly improve the physicochemical characteristics of the films and the corrosion resistance of the substrate [10, 11]. The adhesion of DLC thin film coatings could be improved by surface treatment techniques such as plasma cleaning [12], incorporating interlayers [13], plasma nitriding [14], carburising [15], nitrocarburising [16], or combination of the above‐mentioned methods [17].

Corrosion behaviours of DLC coatings on nitrided steel substrates have been investigated in few studies [11, 18, 19]. Jellesen et al. [18] have shown that corrosion resistance of low‐temperature plasma nitrided SS could be improved after applying a DLC thin film. Azzi et al. [11] investigated the improvement of the corrosion resistance of DLC thin films on plasma nitrided 316L SS with increasing breakdown potential (E _bd) and decreasing corrosion current. In the most recent study, Dalibón et al. [19] have reported that the corrosion resistance of plasma nitrided Corrax SS could be improved by DLC thin films with increasing E _bd, but there was no significant effect on the reduction of corrosion current density.

To the best of our knowledge, the corrosion behaviour resulting from DLC thin films on plasma nitrocarburised (CN‐treated) American Iron and Steel Institute (AISI) 316L SS medical implants has not been studied so far. Therefore, the objective of the present paper was to investigate the physicochemical and biological characteristics of AISI 316L SS after plasma CN treatment and DLC thin film deposition (CN+DLC) for potential bone tissue engineering applications.

2 Experimental results

AISI 316L SS implants were used in dimensions of 10 × 10 × 3 mm³. A pulsed DC plasma‐enhanced chemical vapour deposition (Plasma Fanavar Amin Company‐Iran) process was used to deposit DLC thin films on the substrates. The substrates were polished using silicon carbide papers up to 1 µm diamond paste and then cleaned ultrasonically in acetone and methanol bath before putting them into the deposition chamber (vacuum base pressure of 2 Pa). The substrates were assembled on a cathode fed by a unipolar pulsed DC source, consisting of pulses with maximum 5 kW and 700 V at a frequency of 12 kHz. The DLC thin films were deposited on the untreated and CN‐treated implants. The substrates were first cleaned whit argon plasma with 16 sccm gas flow at 14 Pa working pressure and discharge voltage of 500 V for 20 min. The CN treatment was performed in the presence of a mixture of methane, hydrogen, nitrogen and argon (6:75:25:75). The pre‐treatment voltage, temperature and pressure were 550 V, 460°C and 280 Pa, respectively. The DLC deposition was performed at pressure of 40 Pa with 4:16 methane to argon flow ratio for 2 h. The voltage, frequency, duty‐cycle and temperature of deposition were 500 V, 12 kHz, 30% and 150°C, respectively. The substrates temperatures were monitored by a thermocouple located on the cathode.

The thickness and surface morphology of the films were observed by a field emission scanning electron microscope (SEM) using Te‐Scan Model of FESEM. The Energy Dispersive X‐Ray Spectroscopy analysis was used to identify nitrogen and carbon elements. X‐ray diffraction (XRD, X'Pert‐Pro Multi‐Purpose X‐ray Diffraction system, PANalytical Company) analysis was used to characterise the actual phases of the CN‐treated implants.

The chemical structure of the thin films were analysed by Raman spectroscopy (Almega Thermo Nicolet Dispersive) with 532 nm of a neodymium‐doped yttrium lithium fluoride laser (30 mW). The sp³ fraction in the thin films was calculated by (1)

$$\mathrm{s}{\mathrm{p}}^3\mathrm{content}=0.24-48.9({\omega }_G-0.1580)$$ (1)

where ω_G is the position of G peak (in inverse micrometre unit) [20].

The corrosion measurements were performed in a three‐electrode system test unit with a platinum counter electrode and an Ag/AgCl reference electrode using A263 Edgerton, Germeshausen, and Grier company potentiostat/galvanostat. All the potentiodynamic tests were carried out at 37°C and pH 7.4 in phosphate‐buffered saline (PBS) electrolyte to simulate the human body fluid environment. The composition of the PBS solution used in this paper was 0.14 M NaCl, 1 mM KH₂ PO₄, 3 mM KCl and 10 mM Na₂ HPO₄. For the polarisation tests, the samples were first immersed in the solution for 30 min to stabilise the open‐circuit potential. Then, potentiodynamic polarisation was carried out at the scan rate of +1 mV/s from −1.3 to +1.5 V. After the tests, the polarisation corrosion resistance (R _p) was calculated using (2)

$$R_{\mathrm{p}}=\frac{{\beta }_{\mathrm{a}}{\beta }_{\mathrm{c}}}{2.3\cdot I_{\mathrm{corr}}({\beta }_{\mathrm{a}}+{\beta }_{\mathrm{c}})}$$ (2)

where β _a and β _c are the anodic and cathodic Tafel slopes, respectively, and I _corr is the corrosion current [21]. Since changes in corrosion potential (E _corr) can also give an indication of forming and dissolving of passive layer on the top surface of samples, corrosion potential monitoring was also conducted. By using electrochemical techniques, it is possible to estimate the porosity of the deposited DLC thin films. The porosity (P) can be determined from the measured polarisation resistance by (3) [22]

$$P=\frac{R_{\mathrm{pm}(\mathrm{substrate})}}{R_{\mathrm{p}(\mathrm{coating}-\mathrm{substrate})}}\times {10}^{-\left|\mathrm{\Delta }{E}_{\mathrm{corr}}/\phantom{\mathrm{\Delta }{E}_{\mathrm{corr}}{\beta }_{\mathrm{a}}}{\beta }_{\mathrm{a}}\right|}$$ (3)

where P is the total coating porosity, R _pm is the polarisation resistance of the CN‐treated implants and R _p is the measured polarisation resistance of the coated steel system. ΔE _corr is the difference of the corrosion potential between the coating and the CN‐treated implants and β _a is the anodic Tafel slope of the CN‐treated implants. Also, protective efficiency (P_i ) of the thin films was determined from the polarisation curve by (4)

$$P_i=100\left(1-\frac{i_{\mathrm{corr}}}{i_{\mathrm{corr}}^0}\right)$$ (4)

where i _corr and $i_{\mathrm{corr}}^0$ indicate that the corrosion current densities in the presence and absence of the thin films, respectively [23].

Adult human mesenchymal stem cells (hMSC, Lonza Walkersville Inc., MD, USA) were cultured in a mesenchymal stem cell basal medium (Mesenchymal Stem Cell Growth Medium, Lonza) with other supplements, as recommended by the protocol from Lonza Walkersville. On confluency, the cells were removed from the plate with Clonetics Trypsin‐ethylenediaminetetraacetic acid (Lonza Walkersville). The cells then underwent centrifugation at 300 g for 5 min and were suspended in the growth medium. Viable cell counting was done using Trypan blue dye exclusion assay. Next, the cells were stained using anamine‐reactive, colourless, non‐fluorescent dye that diffuses into the cytoplasm of the cells, 5‐(and‐6)‐carboxyfluoresceindiacetate, succinimidyl ester (CFDA‐SE)‐mixed isomers obtained from Invitrogen Corp. Carlsbad, CA, USA. Cell viability after 24 and 72 h was assessed by two approaches: first, 100 µl of spent media collected was used to analyse cell viability indirectly. Second, the specimens containing the cells were washed in mesenchymal stem cell growth medium, and cytoplasmic CFDA‐SE stained extracted from the live cells by three cycles of repeated freezing and thawing. The CFDA‐SE content in the spent medium and the cytoplasm was assessed by fluorescence intensity in a Gemini Model of Spectrofluorometer spectrofluorometer (Molecular Devices technologies, Santa Clara, CA) at the excitation and emission wavelengths of 485 and 525 nm, respectively. All fluorescence values for the samples were normalised to the control sample for comparison.

3 Results and discussion

All of the deposited DLC thin films on untreated implants were delaminated from the substrates because of poor adhesion of the films to untreated implants. Fig. 1 a shows a typical XRD pattern of a CN‐treated implant. The CN‐treated sample consisted Fe₄ N (γ ′), CrN, Fe₃ C phases according to JCPDS # 06‐0627, 11‐0065 [24] and 35‐0772 [25], respectively. It is worth mentioning that an expanded austenite (γ_N ) phase was also observed in the treated implants [26].

Fig. 1.

Typical XRD pattern of a CN‐treated implant

(a) XRD pattern of the CN‐treated implant, (b) Typical cross‐sectional SEM micrograph of the CN‐treated implant, (c) EDS line scan of carbon and nitrogen elements of the CN‐treated implant

The SEM micrograph of the cross‐section of CN‐treated implant is shown in Fig. 1 b. A distinct boundary interface was found between the compound layer and implant substrate which was about 10 μm in thickness. The energy dispersive x‐ray spectroscopy (EDS) line scan results of carbon and nitrogen elements as shown in Fig. 1 c, demonstrated that the CN treatment produced a dual‐layer microstructure with 3.5 μm carbon and nitrogen‐enriched layer on the top of 7 μm nitrogen‐enriched layer. Marušić et al. [27] have recently analysed the surface of a non‐alloyed carbon steel subjected to thermochemical modification by CN treatment in order to improve mechanical properties, corrosion and wear resistance. They suggested that nitrocarburising process could significantly improve the surface characteristics of the non‐alloyed carbon steel by the diffusion of carbon and nitrogen into the substrate, which is in agreement with the obtained results in our paper.

Raman spectrum obtained from the applied DLC thin films, as shown in Fig. 2 a. Typically, amorphous carbon structures shown common features in their Raman spectra in the range of 1000–2000 cm⁻¹, in which two peaks are usually detected for graphitic D and G bands [28]. Here, Raman spectrum of the DLC thin films demonstrated that D and G band positions are at 1529 and 1386 cm^‐1. It is known that G band is related to sp² carbon bonds in both aromatic chains and rings, whereas D band is related to disorders and breathing modes of sp² atoms in rings [28]. The I (D)/I (G) ratio, sp³ content, G band half peak width calculated from the Raman spectra are also listed in Fig. 2 a.

Fig. 2.

Raman spectrum obtained from the applied DLC thin films

(a) Raman analysis of the CN+DLC thin film with calculated data extracted from the Raman spectra, (b) Typical cross‐sectional SEM micrograph of the CN+DLC thin film, (c) Typical SEM micrograph of the surface morphology of the CN+DLC thin film

A typical SEM micrograph of the cross‐section structure of the thin film coatings is shown in Fig. 2 b. It can be seen that the thickness of the film is about 765 nm with deposition rate of 6.37 nm/min. Fig. 2 c also shows the SEM micrograph of the DLC thin film surface. It can be clearly seen that there is no any detectable porosity in the film surface.

Fig. 3 shows the polarisation curves for the different samples in PBS solution. It can be seen that a passive layer was occurred on the surface of all samples. The measured E _corr, i _corr, P, P_i , E _br, R _p and the current density at the beginning of passivation (i _passive) are given in Table 1. The i _corr and E _corr of the CN‐treated implants were increased and then decreased, respectively, which clearly shows that breakdown potential of CN‐treated samples was decreased in comparison with the untreated implants. Furthermore, the corrosion rate of CN‐treated implants in the passive region was increased. This behaviour was mainly attributed to the presence of Cr nitride and carbide phases in the layer which reduces the corrosion resistance of 316L SS. In another word, this can be also described as a local depletion of Cr and preventing the formation of a continuous passive Cr₂ O₃ layer as well as galvanic cell reaction between Cr nitrides/carbides and Fe₄ N phases [5] due to the difference between their corrosion potentials. The i _corr of CN+DLC samples was reduced by two order of magnitude and by a significant increase in breakdown potential in comparison with untreated and CN‐treated samples. It is also worth mentioning that the corrosion rate of CN+DLC samples in the passive region is about ten times lower than the corrosion rate of the untreated implants in the passive region. Fig. 3 also shows the SEM micrographs of different samples after corrosion test. This clearly indicates that the applied DLC films have excellent pitting and corrosion resistance and no considerable damage observed in comparison with other samples tested. The DLC coatings are known to be chemically inert, so it is expected to corrode at very low rates [29]. The mechanism for the corrosion resistance improvement of CN+DLC is related to low electrical conductivity of the DLC structure which reduces charge transport, and therefore improves the electrochemical properties [30]. On the other hand, according to Fig. 2 c as well as the calculated film porosity, it can be said that the thin films have a very low porosity suitable for biological applications. So, this could be another reason for desirable film corrosion protection as Koskinen et al. [31] have also proved that for an adequate protection against corrosion.‏

Fig. 3.

Potentiodynamic polarisation curves of the untreated, CN‐treated and CN+DLC thin film implants, with the surface observation of corresponding samples after corrosion tests

Table 1.

Results of potentiodynamic polarisation tests

Sample	i corr, A/cm2	E corr, V	E br, V	i passive, A/cm2	β a, V/decade	β c, V/decade	R p, Ω.cm2	Porosity, %	Pi , %
untreated	3.27 × 10−6	−0.221	0.413	6.21 × 10−6	0.54022	0.16168	16545.3	—	—
CN‐treated	6.57 × 10−6	−0.857	−0.312	3.31 × 10−5	0.15518	0.16198	5244.7	—	—
CN+DLC	1.87E −08	−0.493	1.301	1.36E −07	0.15036	0.10952	1468745.6	1.61E−03	99.71

Using pre‐staining the cells with CFDA‐SE, the viability of hMSCs on the samples was evaluated after 24 and 72 h of culture. According to the cytoplasmic contents obtained from the freeze/thaw cycle [see Fig. 4 a], the cell viability for all of the samples was considerable and comparable with the untreated sample, suggesting that the CN+DLC thin films had superior advantages in terms of biocompatibility compared with the other samples. Fig. 4 b also shows that the cell's spreading and morphology in contact with the implants coated with CN+DLC thin films. As can be seen, the cells actively secrete extracellular matrix (ECM) and leave traces of ECM along their migration path. In summary, this paper showed that CN+DLC thin films coating is an efficient surface modification approach to improve the biocompatibility of SSs orthopaedic implants.

Fig. 4.

Cytoplasmic contents obtained from the freeze/thaw cycle

(a) Cell viability on the samples after 24 and 72 h determined by fluorescence in the cytoplasmic extract, (b) Typical optical microscopy image of the CN+DLC thin film after 72 h

4 Conclusion

This study demonstrated that plasma nitrocarburising of 316L SS orthopaedic implants, as a pre‐treatment process, is necessary for the adhesion of DLC thin film coatings. The proposed thin films were a suitable way for the improvement of corrosion resistance by eliminating the formation of Cr nitrides/carbides on the surface of the implants. The DLC thin film coating on the CN‐treated implants significantly improved its corrosion resistance and biocompatibility with respect to the CN‐treated and untreated samples.