1 Introduction

Diamond-like carbon (DLC) films have attracted the attention of many researchers due to their combined properties of graphite and diamond such as high hydrophobicity and hardness, wear resistance, and good optical coefficient among the other characteristics [1,2,3,4,5]. DLC thin films are good candidates for a variety of applications, including a protective cover on anti-reflection coating, solar cells, biomedical devices, and memory storage ones [6,7,8,9,10]. To improve the properties such as optical, mechanical, and so on, many authors have studied the effects of doping DLC films by various elements such as O, Si, B, N, and Ag [11,12,13,14,15]. It is reported that by incorporating nitrogen into DLC coatings, the friction and internal stress of the film can be reduced [16, 17].

In DLC thin films, mechanical properties such as residual stress, depend mainly on the amount of sp3 bonding, which is mainly under the influence of ion energy being equal to the growth parameters and dopant type [18,19,20]. Residual stress, present in DLC thin films, can be reduced via doping by various elements. Amongst different elements, nitrogen has been widely studied as a dopant in DLC thin films, which helps to reduce the stress of DLC thin films [21]. When nitrogen is incorporated into a DLC thin film, C–N groups will take form and operation as an electron receiver due to their strong electron withdrawing capability is done [22]. Kopustinskas et al. studied α-CNx:H films prepared with a ion beam deposition from C6H14 + H2 + N2 gas mixture on Si (100) substrates and observed that, due to increasing nitrogen in the gas mixture, refractive index, growth rate, and sp3 bonding decreased and there was more graphite-like formation [23]. Tsuchiya et al. reported the preparation of nitrogen-doped DLC (N-DLC) films with PECVD technique by CH4, N2, and Ar gases on Si substrate and also investigated the effect of N2 incorporation on their structural, electrical, optical, and mechanical properties. They described that the internal stress of the prepared DLC samples was reduced (less than 0.5 GPa) by N doping at N2 flow rates to less than 13.6% and the electrical resistivity decreased (less 10−2 Ω cm) [24]. Lee et al. reported characteristics of N2 incorporation into DLC thin films deposited with the unbalanced magnetron sputtering for electronic devices. The measured electrical resistivity of their DLC samples reduced while the hardness and elastic modulus of the samples increased with the amount of N2 gas [25]. Jonas et al. reported stability of a-C:N:H layers deposited by RF plasma enhanced CVD method. They described that the temperature rise resulted in the deposition rate decrease. This unfavorable effect may be reduced by application of increasing gas pressure and/or applying higher plasma RF generator power. In addition, when the temperature rise effect is reduced via application of either higher power or increased pressure, the content of nitrogen in the layer increases while the contribution of the CHn terminal bonds decreases [26]. In another article, Jonas et al. reported formation of SixNy(H) and C:N:H layers by plasma-assisted chemical vapor deposition method [27]. They observed the following bands: C–N (1020–1280 cm−1), CN (1610–1660 cm−1), N–H bending (1590 cm−1), NH2 bending (1555–1590 cm−1), CHn bending (1350–1450 cm−1), C=C (1300–1500 cm−1), C=N stretching (2200 cm−1), CHn stretching (2800–3000 cm−1), and N–H stretching (3300–3500 cm−1). Also, with the increase of deposition time, the intensity of all absorption bands characteristics increases. They showed that for all tested samples, the roughness of the layer’s surface increased with deposition process time and the fine granular structure was observed [27]. One of the main applications of DLC thin films is the hardening of lenses and optical windows in the infrared and ultraviolet wavelength regions. Therefore, the main goal of the present study is to investigate the effect of changing the nitrogen flow rate on structural and mechanical properties of the layers. The main novelty of this paper is the preparation of DLC with nitrogen as dopant for first time by direct ion beam technique. This method has some advantages relative to the other ways for preparation of DLC layer such as faster process, ability of using different gases, ability of producing solid diamond carbon layers and industrial samples.

In the present investigation, N-DLC films were deposited by direct ion beam at room temperature on glass and Si substrates. The optical, structural, morphological, and mechanical properties of the prepared samples were studied. Raman analysis was used to investigate the structural properties of N-DLC samples. Morphological properties of the N-DLC samples were studied via atomic force microscopy (AFM). By calculating the curvature of the films after deposition and using the Stoneyʼs equation, the internal stress was measured. The nanoindentation tester was applied to investigate the mechanical properties of the DLC samples. Morphological characteristics showed reduction in film average roughness upon increasing nitrogen flow rate. Raman spectroscopic studies revealed that the amount of graphitic sp2 bonding and size of the graphite cluster increase with N doping. The N-DLC films prepared by direct ion beam deposition in this study showed good optical and mechanical properties indicating their capability of being used in optical devices. The novelty of this paper may be the finding that by increasing the amount of nitrogen content of an N-DLC film, its band gap energy, stress and roughness decreases, while its optical transmittance does not change significantly, so it can be used as a protective layer on the optical lenses.

2 Experimental details

N-DLC samples were prepared using anode layer system by 150 sccm methane (high purity > 99.999%) on the glass and Si (100) substrates with the thickness of 2.05 mm and diameter of 34.91 mm. Figure 1 shows a diagram of anode layer system as well as the working principle. The gas was introduced directly in the channel of discharge. The anode layer system was powered by a DC power supply in the voltage regulation mode. The strong magnetic field between the external and internal poles, in addition to keeping a strong electric field in the discharge plasma, led the electrons to be suspended. Glass and Si (100) substrates were cleaned in an ultrasonic bath with acetone and, then, dried by high purity N2 gas (purity 99.999%). The substrates were fixed on a static holder at an approximate distance of 15 cm from the ion source. The DLC thin films were doped by different nitrogen flow rates (5, 10, 20, 40, and 50 sccm). They were labeled as N5-DLC, N10-DLC, N20-DLC, N40-DLC, and N50-DLC, respectively. The ion source was worked at the voltage of 1.9 kV to clean plasma of the substrate in the chamber before deposition, while argon gas (purity 99.99%) was flown via the gun by the flow rate of 25 sccm. For all the processes, the pressure in the vacuum system pumped down to 6 × 10−6 mbar.

Fig. 1
figure 1

A schematic of anode layer source and working principle

Full size image

Surface morphology of all the N-DLC samples was investigated by AFM (Park Scientific AUTO PROBECP). Their optical properties were further examined by a UV–Vis spectrophotometric (Perkin-Elmer). The structural properties of the prepared specimens were also explored by Raman analysis (Takram P50C0R10 model: power up 0.5 to 70 mW and the excitation wavelength of 532 nm). The internal stress was calculated according to the Stoney equation from the curvature radius of the substrates measured by an optical interferometry. Finally, the mechanical features of N5-DLC and N40-DLC samples were evaluated by a nanoindentation tester (Compact Platform, Switzerland) using the continuous stiffness measurement (CSM) technique. The CSM technique is the most reliable method which has been widely used for evaluating the mechanical properties of thin films from the load–displacement curves of the indenter (CSM Instruments).

3 Results and discussion

3.1 Morphological characterization

Field emission scanning electron microscope (FESEM) was used to measure the thickness of the DLC thin films. Figure 2 represents the cross section and surface views of the DLC thin films deposited at different nitrogen flow rates (0, 5, 10, 20, 40, and 50 sccm). The thickness of DLC thin films obtained after processing by using Digimizer software [28]. The thickness of DLC, N5-DLC, N10-DLC, N20-DLC, N40-DLC, and N50-DLC samples were estimated 269, 285, 250, 274, 290, and 260 nm, respectively. The cross section images indicated that all the samples were uniform, compact with good adhesion on glass substrate, the reason for which may be attributed to the use of methane or propane as precursor of carbon [29,30,31]. Furthermore, good agreement was found between actual and estimated thicknesses of the samples. Comparing the surface images of the samples show that with increasing nitrogen flow rates, nanoscale particles has increased. These nanoscale particles are more specific in N10-DLC, N20-DLC, and N50-DLC samples. These results are in good agreement with the results obtained using AFM analyze.

Fig. 2
figure 2

The cross section and surface views of the DLC thin films deposited at different nitrogen flow rates

Full size image

Morphological characteristics of the N-DLC samples were studied via AFM analysis. Figure 3 indicates the 3D AFM images of DLC, N5-DLC, N10-DLC, N20-DLC, N40-DLC, and N50-DLC samples. By increasing the N2 content from 5 to 40 sccm, their average roughness reduced and their surface became much smoother. The average roughness (Ra) of DLC film alone was estimated 4.3 nm. The average roughness (Ra) as a function of N2 incorporation in the DLC film is indicated in Fig. 4. With the increment of N2 incorporation (from 0 to 40 sccm), the average roughness reduced (from 4.5 to 0.68 nm). Doping nitrogen into DLC films decreased their hydrogen content [32]. Generally, the surface roughness of DLC film is affected by surface motion, graphitic recrystallization, and substrate temperature [33]. In the present study, the substrate temperature was kept constant for all the samples, while graphitic recrystallization changed for different N2 contents. The root-mean-square (RMS) roughness of all the N-DLC ranged from 2.7 to 1.02 nm. The RMS roughness values as a function of N2 incorporation are indicated in Fig. 4. Tang et al. [34] reported RMS values of 0.35 nm for α-C:H:N and 0.5–0.8 nm for α-C:H samples. The RMS values depend on the deposition method and the thickness of the layers. Comparing the results, it can be found that End-Hall ion beam deposition produces smoother DLC thin films than direct ion beam method even if the films prepared in this work are much thicker. Also RMS and surface roughness increase linearly with film thickness, as reported by Logothetidis and Stergioudis [35]. Waltman et al. [36] reported RMS roughness values of 0.8 and 1.2 nm for α-C:H:N and α-C:H samples. All parameters for the morphological characteristics of the DLC samples are given in Table 1. The thickness of DLC samples was determined using FESEM and the values are given in Table 1.

Fig. 3
figure 3

3D AFM images of all the DLC samples

Full size image
Fig. 4
figure 4

The Ra values and RMS values versus of N2 flow rate

Full size image
Table 1 Optical, structural, mechanical, and morphological parameters of N2-DLC samples at various N2 contents
Full size table

3.2 Raman characterization

The DLC films bonding configuration was analyzed by Raman spectroscopy. Raman analysis is a standard tool to describe the microstructural information and the nature of carbon bonding [37]. Raman analysis of DLC films is usually composed of two peaks: D (disorder) peak around wavenumber 1370 cm−1 and G (graphite) peak situated in the wavenumber range 1500–1650 cm−1. The formation of ordered sp2 clusters and sp2 vibrations of stretching caused by the aromatic ring, respectively [38, 39]. Raman plot of all the nitrogen doped samples is shown in Fig. 5. In this figure, two Raman peaks of 1576 cm−1 and 1367 cm−1 related to G and D peaks are respectively observed, indicating a typical DLC structure. Figure 6 demonstrates fitting two Gaussian peaks of N5-DLC, N20-DLC, and N40-DLC thin films. One can basically observe G peak movement to the higher values and increment in the intensity of D peak upon increasing N2 gas flow rate in the figure. From fitting data, the intensity ratio of D and G peaks (ID/IG), G peak position, and full width at half maximum (FWHM) of peaks can be extracted. The G peak position and FWHM of the DLC samples versus N2 gas flow rate are indicated in Fig. 7. It is observed from the figure that by increasing N2 content, G peak position moved to the higher wavenumbers. It was also observed that DLC thin films were graphitized, which could be due to the growth and formation of sp2 clusters [40]. These behaviors were attributed to the structural ordering of DLC samples [38]. Lee et al. [38] reported that the G peak position moved to the higher intensities by increasing N2 content. Tsuchiya et al. [24] prepared N-DLC films by PECVD technique and studied the effects of N content on the structural properties. They indicated that the G peak moved to the higher wavenumbers by increasing N2 content, which could be due to the structural ordering of DLC films.

Fig. 5
figure 5

Raman plot of all the nitrogen doped DLC samples

Full size image
Fig. 6
figure 6

Fitted two Gaussian peaks for N5-DLC, N20-DLC, and N40-DLC thin films

Full size image
Fig. 7
figure 7

G peak position and FWHM of all DLC samples

Full size image

By increasing N2 content, FWHM of G peak shifted to the lower wavenumbers. As the sp2-rich cluster size with the N2 content increased, the disorders in the DLC samples diminished and resulted in a reduction in the FWHM of G peak [18]. Polaki et al. [18] reported DLC samples deposited on Si (100) substrate with different N2 contents by PECVD technique. By increasing N2 flow rate, they indicated a structure of graphitic in DLC samples.

It is necessary to discuss about the intensity ratio of D and G peaks (ID/IG) in order to describe the structure of the DLC samples. The intensity ratio ID/IG is a measure of the number and size of sp2 clusters organized in rings. High ID/IG ratio of the DLC samples is an indication of increment in the sp2 clusters in aromatic rings, whereas its low ratio means the sp2 clusters are relatively organized in chains. The ID/IG ratio of all the N-DLC thin films versus N2 content is represented in Fig. 8. It can be observed that this ratio sharply increases by N2 content. The size of the graphite cluster (La) was calculated using the Ferrari and Robertson equation [41]:

$$\frac{{{\text{I}}_{\text{D}} }}{{{\text{I}}_{\text{G}} }} = {\text{C}} \cdot {\text{L}}_{\text{a}}^{2} ,$$
(1)

where C is ∼ 0.0055. La values versus N2 gas flow rate for all the N-DLC samples are presented in Fig. 8. It can be seen that the size of graphite clusters increased with the N2 content. The ID/IG ratio also increased from 0.74 to 0.9, followed by an increment in the graphite cluster size from 1.16 to 1.28 Å. All the parameters regarding the structure of the DLC samples are given in Table 1.

Fig. 8
figure 8

The ID/IG ratio and La of all the nitrogen doped DLC samples

Full size image

3.3 FTIR measurement

FTIR plot of all the N-DLC samples on Si substrates in the range 1400–3500 cm−1 is presented in Fig. 9. In this figure, a few peaks belonging to the FTIR characteristics of DLC film doped with N2 gas, are observed. The figure indicates a broad peak in FTIR pattern of all the samples at 2910 cm−1 due to C–H stretching vibrations mode. A reduction in the intensity of the C–H bond by increasing the N2 doped is observed, showing low hydrogen incorporation in the N50-DLC samples. Another group of intense absorption band at around 2400 cm−1 is attributed to C–O bonds. An additional group of intensive absorption band observed at around 2420 cm−1 is relevant to C–O [41, 42]. The intense absorption band centered at 1450 cm−1 is identified as the stretching vibration of C=C bond, which is in agreement with the reported result [43]. Considering Fig. 9, three peaks including C–N, C≡N, and N–H mods appeared at 1740 cm−1, 2330 cm−1, and 3360 cm−1, respectively. Our analysis of FTIR data matches with the results reported by others [44, 45]. The research conducted by Kaufman et al. [44] shows two lines active in the sp2 area at around 1550 and 1375 cm−1. By increasing N2 incorporation, a weaker doublet was indicated at 2110–2210 cm−1 in the sp area [44]. The 2850 and 2958 cm−1 peaks are assigned to the symmetric modes of sp3-CH3 and sp3-CH2 groups, respectively [45].

Fig. 9
figure 9

FTIR plot for all the nitrogen doped DLC samples in the range 1400–3500 cm−1

Full size image

3.4 UV–Vis analyzes

The transmittance and reflection of N-DLC samples prepared at different N2 flow rates is demonstrated in Fig. 10. The average optical transmittance of N5-DLC, N10-DLC, N20-DLC, N40-DLC, and N50-DLC samples in the 400–2200 nm wavelength region was calculated as 67.88, 66.67, 61.09, 57.70, and 61.87%, respectively. This result indicates that the optical transmittance of N-DLC films diminished by increasing the N2 flow rate. This finding is evidence of the microstructure changes in the samples as the result of increasing the N2 content. This reduction in optical transmittance is probably related to the growth of the sp2 fraction.

Fig. 10
figure 10

The transmittance of N-DLC samples with various N2 flow rate in the UV–Vis–NIR area

Full size image

The absorption coefficient (α) has been measured by the following equation [46] to get the optical band gap (Eg) of N-DLC samples:

$$\alpha = \frac{1}{d}\ln \frac{{\left( {1 - R} \right)^{2} }}{T},$$
(2)

where T is the transmission, R is the reflection, and d is the thickness of the films. The optical band gap of the samples has been calculated using Tauc equation [47]:

$$\left( {\alpha h\upsilon } \right)^{1/2} = A\left( {h\upsilon - E_{g} } \right),$$
(3)

where h is the Planck’s constant, A depends on the effective masses of holes and electrons and is constant, and n is a constant equal to 0.5 for direct band gap as well as 2 for an indirect optical band gap. Figure 11 shows the plot of \(( \alpha h\upsilon )^{1/2}\) as a function energy of photons (hυ) in N-DLC samples prepared at various N2 flow rates. The band gap energy values are obtained by extrapolating the linear portion of the curves to x axis. The results are presented in Table 1. The estimated optical band gaps for the DLC, N5-DLC, N10-DLC, N20-DLC, N40-DLC, and N50-DLC samples were 2.87, 2.74, 2.48, 2.38, 2.2, and 2.29 eV, respectively. This results shows that when the N2 flow ratio was increased the optical bandgap decreased. Rodil and Muhl [48] showed that the bandgap of tetrahedral carbon nitride films decreased sharply with increasing N/C ratio and the opposite trend was observed for amorphous carbon nitride films. Tsuchiya et al. [24] showed that when the N2 flow ratio was increased from 0 to 10%, the optical bandgap decreased from 1.16 to 1.06 eV. Zkria et al. [49] showed that the indirect optical bandgap decreases from 1.7 to 1.28 eV by nitrogen doping. Seker et al. [50] reported the optical band gap for N-DLC samples in the range of 1.92 to 2.29 eV. Also, they reported that the optical band gap decreased upon increasing N2 contents. It is clear that N2 contents favored the increase of sp2 clusters in size and number [21], which was consistent with the results of Raman studies already provided. Considering the results, the increase in the disordering and size of sp2 clusters would affect the optical properties of N-DLC samples. The band gap of DLC samples was not only proportional to the sp3 sits but controlled using distortions and ordering of π states in sp2 phases [51, 52]. The π state of sp2 phases was near the Fermi level conduction and valence band edges. The reduction in the energy gap sourced from the \(\pi /\pi^{*}\) transfer occurred in the sp2 domain [53].

Fig. 11
figure 11

The curve of \(\left( {\alpha h\upsilon } \right)^{1/2}\) as a function hυ of N-DLC samples with various N2 flow rates

Full size image

3.5 Evaluation of internal stress

To calculate the internal stress of all the samples, slide glass substrate (34.91 mm in diameter and 2.05 mm in thickness) was selected. The internal stress was measured by calculating the film curvature after deposition on one side of the substrate via the Stoneyʼs relation [54]:

$$\sigma = \frac{{E_{s} }}{{6\left( {1 - \nu_{s} } \right)}}\left( {\frac{{t_{s}^{2} }}{{t_{f} }}} \right)\left( {\frac{1}{{R_{2} }} - \frac{1}{{R_{1} }}} \right),$$
(4)

where \({{\sigma }}\) is the internal stress, \(\nu_{s}\) is the substrate Poisson’s ratio, \(E_{s}\) is the substrate Young modulus, R1 and R2 are the substrate curvatures before and after deposition, ts and tf are the thicknesses of substrate and film, respectively. The values adopted for the glass substrate constants (\(\nu_{s}\) = 0.22, \(E_{s}\) = 1.03 × 1010 Nm−2) were taken from Zou et al. [55]. The internal stress of N-DLC samples was calculated by the stress induced from the interferometric surface profiler. The radius (R) of the substrates before and after the deposition was calculated using the observation of Newtonʼs rings method by the optical interferometry. The radius of the substrate curvature was determined using the following equation:

$$R = \frac{{d_{m}^{2} }}{4m\lambda }\quad \left( {m = 1,\;2,\;3, \ldots } \right),$$
(5)

where dm is the mth dark interference fringe diameter and \(\lambda\) is the wavelength of the light (\(\lambda\) = 589.3 nm). Figure 12 indicates the internal stress of N-DLC samples versus N2 content. It was observed that, with the N2 content increase, the internal stress declined. By increasing the N2 content from 0 to 40 sccm, the internal stress reduced from 1.9 to 0.58 GPa. For the N50-DLC sample, the internal stress was measured as 1.4 GPa. Sharifahmadian et al. reported the same results for N-DLC thin films [56]. The three important factors causing the reduction in internal stress are increment of the sp2-C bonds [57], increase of the induced ad-atom mobility [58], and distortion of the bond angle [58]. Raman analysis confirmed that incorporating N2 into DLC films increased the values of sp2 hybridized atoms. In addition, the incorporation of N2 decreased the coordination numbers of the films, which in turn declined the degree of over-constraining in the N-DLC samples; hence decreasing the stress [59]. Furthermore, by the N2 incorporation, C≡N bonds were formed, leading to a weaker network between the sp2 clusters in the N-DLC films, also reducing the internal stress in the films [60]. As illustrated in Fig. 12, the internal stress of N50-DLC was higher than that of N40-DLC sample. The higher internal stress of the N50-DLC film might be due to the organization of a larger amount of N2 plasma during its deposition process [21].

Fig. 12
figure 12

The internal mechanical stress of N-DLC samples versus N2 content

Full size image

3.6 Mechanical properties

Finally, nanoindentation measurements were done on N5-DLC and N40-DLC samples to investigate their mechanical properties. Hardness (H), elastic modulus (E), plasticity (H/E), residual displacement after load removal with displacement at the most load (hres/hmax), and elastic recovery (%ER) were performed following Oliver and Pharr’s model [61]. The maximum depth in the nanoindentation test was limited to lesser than one-tenth of the thickness of the DLC sample. The load versus displacement plot for N5-DLC and N20-DLC samples is shown in Fig. 13. The results of H and E for these samples are indicated in Table 2. Mechanical hardness was measured from 5 points of the sample surface and the values given in Table 2 are the average mechanical hardness. According to the results, high hardness and elastic modulus were obtained for the samples. It was found that with increasing N content from 5 to 40 sccm, hardness was declined from 26.4 to 22.6 GPa, in addition, elastic modulus was decreased from 107.9 to 104.3 GPa. It may be mentioned that E and H of DLC samples depend on the sp3 bonding fraction [62]. π Bonds of sp2 bonding structures lead to the reduction of the mechanical properties and sp3 bonding is combined with a covalent bond to improve the mechanical properties. According to the Raman’s analysis, sp2 bands increases by the increase of nitrogen content; thus, it may be the reason for a decline in the hardening and the elastic modulus of the samples. This observation was very similar to those reported by others [63, 64].

Fig. 13
figure 13

Load–displacement curves of the N5-DLC and N40-DLC samples

Full size image
Table 2 Values of H, E, %ER, H/E, and hres/hmax for N5-DLC and N40-DLC films
Full size table

The plasticity index parameter (H/E) is the main parameter in explaining the wear resistance properties of films due to combined effect [65]. The H/E values for N5-DLC and N40-DLC films are presented in Table 2. The results show that by increasing nitrogen content, the H/E decreased from 0.244 to 0.216, because the graphite-like sp2 bonding formation started. The %ER is the main parameter for hard coatings. Higher the %ER factor value higher be the hardness value of the films and more useful the coatings for the tribological applications [65]. The values of %ER were measured with the following equation [66]:

$$\% {\text{ER}} = \frac{{{\text{h}}_{\hbox{max} } - {\text{h}}_{\text{res}} }}{{{\text{h}}_{\hbox{max} } }} \times 100.$$
(6)

As can be seen in Table 2, %ER increased from about 69% to 86% by increasing N incorporation from 5 to 40 sccm. The results of %ER are in good agreement with the similar results for H. Finally, the hres/hmax parameter provided information analogous to that of ER with a variable domain of validity. The values of hres/hmax for N5-DLC and N40-DLC samples are presented in Table 2. Referring to Table 2, the values %ER decreased from 69 to 86% by increasing N content from 5 to 40 sccm. It was also found that the hres/hmax values were in the line with H, E, and H/E results.

4 Conclusions

N-DLC films were prepared by direct ion beam deposition technique (at room temperature) on Si and glass substrates to study the effects of N doping on their structural, optical, and mechanical properties. Morphological characteristics showed a reduction in average film roughness (from 4.1 to 0.68 nm) upon increasing nitrogen flow rate (from 5 to 40 sccm). Raman studies revealed that the number of graphitic sp2 bonding and the size of the graphite clusters increased from 1.16 to 1.27 Å with N doping. With increasing nitrogen flow rate (from 5 to 40 sccm) the energy gap decreased from 2.61 to 2.27 eV and the internal stress of the deposited samples decreased from 1.63 to 0.58 GPa. The mechanical features of the DLC films show that with increasing N incorporation from 5 to 40 sccm, hardness declined from 26.4 to 22.6 GPa, elastic modulus also decreased from 107.9 to 104.3 GPa. Therefore, in this study, the N-DLC samples prepared by direct ion beam deposition technique showed good optical and mechanical properties; indicating the possibility of using them in optical devices.