Effect of surface structure and wettability of DLC and N-DLC thin films on adsorption of glycine

Applied Surface Science 258 (2012) 5166–5174

Contents lists available at SciVerse ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Effect of surface structure and wettability of DLC and N-DLC thin films on adsorption of glycine Mukhtar H. Ahmed ∗ , John A. Byrne Nanotechnology Integrated Bio-Engineering Centre, University of Ulster, Jordanstown, BT37 0QB, Belfast, UK

a r t i c l e

i n f o

Article history: Received 29 November 2011 Received in revised form 4 January 2012 Accepted 29 January 2012 Available online 9 February 2012 Keywords: Diamond-like carbon (DLC) Wettability Surface free energy Glycine adsorption Raman spectroscopy Fourier transform infrared spectroscopy (FTIR)

a b s t r a c t Diamond-like carbon (DLC) is known to have excellent biocompatibility. Various samples of DLC and nitrogen-doped DLC thin films (N-DLC) were deposited onto silicon substrates using plasma-enhanced chemical vapour deposition (PECVD). Subsequently, the adsorption of amino acid glycine onto the surfaces of the thin films was investigated to elucidate the mechanisms involved in protein adhesion. The physicochemical characteristics of the surfaces, before and after adsorption of glycine, were investigated using Fourier transfer infrared (FTIR), Raman spectroscopy, spectroscopic ellipsometry (SE) and contact angle (). The Raman study highlighted decrease slightly in the ID/IG ratio at low levels of N (5.4 at.%), whilst increasing the nitrogen dopant level (>5.4 at.%) resulted in a increase of the ID/IG ratio, and the FTIR band at related to C N. Following exposure to glycine solutions, the presence of Raman bands at 1727 cm−1 and 1200 cm−1 , and FTIR bands at 1735 cm−1 indicates that the adsorption of glycine onto the surfaces has taken place. These results which obtained from SE and surface free energy, show that low levels of nitrogen doping in DLC enhances the adsorption of the amino acid, while, increased doping led to a reduced adsorption, as compared to undoped DLC. Glycine is bound to the surface of the DLC films via both de-protonated carboxyl and protonated amino groups while, in the case of N-DLC gylcine was bound to the surface via anionic carboxyl groups and the amino group did not interact strongly with the surface. Doping of DLC may allow control of protein adsorption to the surface. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The implantation of biomaterials into the human body allows it restructure function and hence to enhance the quality of life. The highly corrosive surroundings and the low tolerance of the body to some dissolution products restrict the materials to be used for implants [1]. Diamond-like carbon (DLC) is an excellent candidate for use as biocompatible coatings on biomedical implants [2] such as rotary blood pumps [3], artificial hearts, mechanical heart valves [4–6], coronary artery stents [7,8], hip and knee replacements [9,10], due to its remarkable properties such as high mechanical properties, high wear resistance, and chemical inertness [11,12]. Comparative studies showed that DLC has better biocompatibility and wear resistance than stainless steel [13], titanium and titanium alloys [14], poly-methyl mehtacrylate (PMMA) [15], cobalt chrome alloys, and alumina ceramics [16]. Several studies were performed to observe the dependence of hemocompatibility on the Raman D-band to G-band intensity ratio (ID/IG) of the DLC films [17]. Therefore, the structure of DLC films plays a vital role on the platelet adhesion on DLC surfaces [18].

In order to improve of it’s properties, DLC films elementally modified by addition of third elements, such as nitrogen, silicon, fluorine, oxygen, and titanium [19,20]. Furthermore, nitrogen doped DLC films (N-DLC) are considered for widespread clinical use as biocompatible coatings due to their excellent mechanical properties including; surface roughness, elasticity, high hardness, infrared transparency and low friction coefficient [21,22]. It was found that hydrogen content of the DLC film as well as the ratio of sp2 to sp3 bonds can have significant effects on friction and wear [23]. The replacement of CH with NH bonds in N-DLC reduces the average coordination number and enhances the sp2 hybridise bonding, leading to decrease in both internal stresses and the sp3 hybridization fraction, due to presence of C N bonds [24]. Investigations have found that the factors including nitrogen concentration, C N film roughness and types of bond between C and N, play a significant role in clotting time and amount of adhered platelets [25].

2. Experimental details 2.1. Film deposition

∗ Corresponding author. E-mail address: [email protected] (M.H. Ahmed). 0169-4332/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2012.01.162

Prior to film deposition, Silicon wafers 1.5 cm × 1.5 cm were washed ultrasonically in pure acetone to remove residual organic

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

5167

Table 1 Deposition of DLC film preparation process parameters employed for the PECVD. Parameters

rf. power (W) Ar:C2 H2 ratio (sccm) Gas flow rate (Torr) × 10−2 N2 gas flow (sccm) at.% N (relative to C and O) [26] Film thickness (nm) by SE

Samples DLC

NI

NII

NIII

109 10:20 0.71 0 0 170 ± 18

117 10:20 1.2 3 5.3 177 ± 16

123 10:20 1.4 5 8.4 188 ± 21

131 10:20 2.0 7 12.1 195 ± 23

Key: bias voltage 400 V, deposition time 5 min, sccm: standard centimetre cube per minute, (±) is SD for (n = 5 samples), at.% N: atomic percentage of nitrogen that were obtained in previous work [26], nm: nanometre (10−9 m), SE: spectroscopic ellipsometry.

contaminants followed washed with distilled water and then dried using a lint-free cloth in a flow of nitrogen gas. DLC and N-DLC films were deposited on substrates by the radio frequency 13.56 MHz plasma enhanced chemical vapour deposition (rf-PECVD) using a Diavac model 320PA (ACM Ltd.), with negative electrode self bias voltages set at 400 V. The experimental equipment had been described previously in details [26]. The films were prepared under the following conditions: discharge power 109–130 W, the source gases used in present experiment was C2 H4 while Argon (Ar) was used as a carrier gas, and nitrogen (N2 ) gas was used as dopant (in case of N-DLC). The C2 H2 /Argon gas mixture flow rate was 20/10 sccm. The N2 gas flow was varied from 0 to 7 ml min−1 Table 1. Prior to deposition, the vacuum chamber was evacuated to ∼7.5 × 10−6 Torr. Glycine (Sigma–Aldrich) was prepared in aqueous phosphate buffer saline (SIGMA) to give solution of 0.001 M at pH 7.4. The DLC and N-DLC coated samples were immersed in a sealed bottle containing of 25 ml of solution at room temperature for 6 h in a shaker. Following incubation, the samples were washed twice with double distilled water and dried in a flow of nitrogen gas.

3. Film characterisation DLC and N-DLC samples were characterised using Raman FTIR spectra, spectroscopic ellipsometry (SE) and contact angle, before and after attachment of glycine. Raman spectroscopy was recorded on a ISA lab-ram model system using an Argon laser beam ∼50 mW 633 nm laser diode for excitation. Prior to acquisition the spectrometer was calibrated using the zero order diffraction peak and first order peak from a silicon phonon mode from a silicon wafer sample. In this work the following parameters, confocal aperture 200 ␮m, spectral resolution 5 cm−1 , A 100× objective was employed and typical acquisition times were 5 s with 7 time repeat. This process was repeated at five different spots across the samples of DLC and N-DLC to assess uniformity of response. The chemical bonding configurations were characterised by Fourier-transformed infrared spectrometer (FTIR). The analysis was performed at room temperature using BIORAD Excalibur (FTS 3000MX series) instrument, the spectrum was recorded in the region of 4000–400 cm−1 and 60 scans were accumulated at a resolution of 4 cm−1 . In all cases of FTIR the background spectrum was collected before the actual sample analysis using the relevant unattached coated samples with glycine and this signature was subsequently removed from the sample scan. The subsequent analysis of the attached glycine on the diamond thin film sample was performed. Static contact angle were performed on the prepared DLC and N-DLC prior to and after glycine adsorption. The degree of wettability of the films was examined with standard solvents; distilled water (H2 O), diiodomethane (C2 I2 ) and ethylene glycol C2 H4 (OH)2 ) with known surface tensions (Table 2), and the sessile drop method using a (CAM 200 optical contact angle system (KSV instruments

LTD, Finland). Drops of 5 ␮L of solvents were generated with a micrometric syringe and deposited on the substrate surfaces. The contact angle was read from a protector of the equipment through a microscope by a naked eye at five different places of each sample surfaces, and the values were averaged. The surface energy ( s ) and measure of the degree of hydrophobcity/hydrophilicty (Giwi ) were calculated according to the equation in references [27,28]. Film thicknesses were performed using spectroscopic ellipsometry (SE) (SOPRA GES-5E) in a room temperature (23 ± 2) ◦ C. The angle of incidence ϕ was set to 68.0◦ and the laser wavelength () was 532.8 nm. The refractive index (nk) and thickness (dk) of prepared samples before and after exposure of glycine can be calculated from the ellipsometric angles,  and  , using the fundamental equation: ei tan = F =R=

Rp = f (nk, dk, , ϕ) Rs

Rp = tan · exp(i) Rs

(1) (2)

where Rp and Rs are the complex overall reflection coefficients for the parallel and perpendicular polarised waves, respectively. They are a function of the angle of incidence ϕ, the radiation wavelength , the indices of refraction, and the thickness of each layer of the model, nk, dk. The ellipsometric quantities (tan  ) [relative amplitude ratio] and (cos ) [relative phase shift] can be derived from positions of the polariser and analyzer at the detector, which are related to the Fresnel reflection coefficients (R) for (p) and (s) polarised light [29]. 4. Results and discussion 4.1. Film characterisation before attachment of glycine 4.1.1. Film thicknesses The film thicknesses of DLC and N-DLC samples have been achieved, and the values were arranged from (170–200) nm, the growth rate of film deposition was around (36 ± 6) nm min−1 , Table 1. 5. Raman spectroscopy The Raman spectra of DLC and N-DLC samples are shown in Fig. 1, in order to obtain graphite (G) and disorder (D) peaks which correspond to sp2 hybridization. The G-peak is due to the bond stretching of all pairs of sp2 configuration atoms in both rings and chains, whilst the D-peak is the shoulder of the G-peak at lower wave-numbers, and is due to the presence of the sp2 aromatic rings [30]. A typical broad peak for the DLC has been observed in the range of 1000–1800 cm−1 , and it was deconvoluted to yield two component peaks at 1349 cm−1 and 1538 cm−1 by Gaussian curve fitting, can be attributed to disordered graphite (D peak) and pure graphite

5168

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

Table 2 Surface tension parameters for solvents used in the study [28]. Solvent

 s mJ m−2

(SD )mJ m−2

(SP )mJ m−2

 + mJ m−2

 − mJ m−2

H2 O CH2 I2 C2 H4 (OH)2

72.8 50.8 48.0

21.8 50.8 29.0

51.0 0 19.0

25.5 0 1.9

25.5 0 47.0

G-band

sp2 (G peak), respectively. For the NI sample, a slight shift in Dband with unvarying value of G-band was observed. Whilst both the G and D peaks were shifted toward higher wave-number with increasing of nitrogen content in coated films at 1543 cm−1 and 1547 cm−1 (G band), with 1357 cm−1 and 1362 cm−1 in corresponding D band for both NII and NIII, respectively (Figs. 1 and 2). The shift band is an indication of the state of development of the sp2 phase and that the sp2 sites are beginning to organise into small graphitic clusters [32]. Nitrogen content in N-DLC samples has also been studied and more presented in a companion paper, and the results showed the slight increase of sp3 /sp2 ratio is observed on goes from 0 to (5.4 at.%) of nitrogen, and then this ratio decreases with increasing of nitrogen concentration in DLC film [26]. It is observed that the ID/IG ratio appears to decrease slightly with NI film type (0.55) as compared to the undoped DLC (0.58). Whilst

800

D-band

Intensity (a/u)

NIII NII NI DLC

1000

1200

1400

1600

1800

2000

ID/IG values increase with increasing of nitrogen doping content i.e.; to (0.748) and (0.815) for NII and NIII, respectively (Fig. 2). The results correspond with reporters who investigated that the ID/IG ratio increases gradually with increasing nitrogen flow rate [33], and these are explained by the formation of more C C bonds which leads to increase of sp2 carbon content in DLC [34]. The results indicate that unsaturated carbon atom in the film can undergo chemical reactions that lead to adhesion between the counterface and the film, increasing the friction, low-hydrogen content with lower density and hardness [23,31]. 5.1.1. Fourier transfer infrared (FTIR) The peaks at 1455 cm−1 assigned to asymmetric (CH3 ) with bands at 1213 cm−1 and 891 cm−1 are assigned to the bending of sp3 hybridised of (C H) bond, Fig. 3, [35,36]. This indicates the formation of sp3 bonded carbon as in the DLC films. This is in agreement with Baek et al., who investigated the DLC films prepared with C2 H2 plasma can be modified to tetrahedral structures by applying the bias volts with the lower frequency [37]. The sp2 hybridised C H stretching appear in the range of 3150–3300 cm−1 , and the band intensity well increased with increasing of nitrogen content. Peak at ∼723 cm−1 is attributed to the out of plane bending vibration of graphite like carbon [8]. A weak band at 1624 cm−1 is assigned to sp2 hybridised (C C). The peak at 2871 cm−1 belongs to the different C H aliphatic group stretching modes [7]. A weak signal at around 1720 cm−1 has been also observed, can be attributed to the presence of a small amount of carbonyl group (C O). On the other hand, in addition of the mentioned above, peaks were observed in N-DLC samples related to nitrogen content in film conformation. The band in range of 1500–1800 cm−1 could be attributed to C N and (C C) bonds [38]. The peak at 707 cm−1 belongs to the out of plane bending configuration resulting from the graphite like domains with nitrogen atoms incorporated [39]. The absorption in range (1100–1300) cm−1 can be attributed to a vibrational stretching mode of C N [40].

Raman shift cm-1 5.1.2. Contact angle and surface energy Fig. 1. Raman spectra of DLC and N-DLC films.

1545

0.85 D band

G band

ID/IG 0.80 0.75

1535

0.70

1360

0.65 0.60

1355

0.55 1350 0

3

6

9

12

atom % of nitrogen in DLC film Fig. 2. Effect of nitrogen atom % on the D and G band positions.

(ID/IG)

Band position (cm -1 )

1540

Fig. 4, shows the mean values of water, CH2 I2 and C2 H4 (OH)2 contact angles ( ◦ ) on the DLC and N-DLC surfaces with corresponding standard deviations. The measured water contact angle of DLC surface is 78.4◦ (±1.9), it is therefore considered as a mildly hydrophobic component. The result is in agreement with those measured by by Yu et al. [41]. For N-DLC, the values of H2 O () are closely related to the atomic concentration ratio of nitrogen to carbon, and it is observed that the water contact angle decreased with increasing of N content in the DLC. As shown in Fig. 4, H2 O () of NI, NII and NIII is 70.3◦ (±1.8), 69.4◦ (±1.9) and 68.3◦ (±1.6), respectively. The N-DLC surfaces exhibit more hydrophilic than undoped DLC film [42]. On the other hand the surface energy of samples increased with increasing of nitrogen content i.e. 49.5, 50.0 and 51.5 mJ m−2 for NI, NII and NIII, respectively (Table 3). This might p be due to an increase of the dispersive sd and polar s components of surface energy [43]. These phenomena related to formation of N H and C N bond which leads to produce high number of nucleophilic electrons. This would stimulate the electronegativity of the surface due to present of pair of non-bonding and (␲) bonding

ar.

5169

C-H

C=N

str. C=C

C-C

C-N

C-H

bend. C-C

C=C

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

(C)

(B)

(A) 400

600

800

1000

1200

1400

1600

1800 2600

2800

3000

3200

3400

3600

wavenumber (cm-1) Fig. 3. FTIR spectrum for DLC (A), NI (B), NII (C) and NIII (D).

5.2. Hydrophobicity of samples (Giwi )

80

Contact angle (θ)

75 (H2O)

70 45 40

(CH2I2)

35 30 25

(C2H6O2)

20

DLC

NI

NII

NIII

Samples Fig. 4. Water, ethylene glycol and diiodomethan contact angle of samples.

electrons in the N-DLC films [44,45]. On the other hand formation of un-saturated double bonds leads to reduction of intermolecular distance and the film density increases, because of ␲ bond ˚ is shorter than ␴ bond (1.54 A). ˚ The later would enhance (1.42 A) the dispersive force in the surface energy of N-DLC films [46]. The electron donor ␥ − value increased with increasing N content which enhanced the polarity of the surface [43,47].

In case of N-DLC, the value of Giwi is generally significantly higher than the value that can be obtained from DLC film (−55.2 mJ m−2 ) Table 3. This indicates that N-DLC has higher hydrophilicity than DLC, and this value increased with increasing of nitrogen content. 5.2.1. Spectroscopic ellipsometry The dielectric function can be obtained from Eqs. (1) and (2). The differences between the dielectric function (ε) values of doped and non-doped DLC samples are very pronounced, because the optical properties (or dielectric function) of DLC films strongly dependent on the ratio of sp2 (graphite like) to sp3 (diamond like) bonded carbon atoms [48,49]. From the analysis date of ellipsometric spectroscopy of DLC film Fig. 5, the ␧ spectrum demonstrates a strong band close to 2.0 eV with broad hump at around 3.15 eV which, is associated with the ␲ → ␲* electronic absorption that takes place due to the sp2 configuration [49]. Whilst for the N-DLC films, the bands observed at around 2.1 eV and 3.8 eV, and these peaks tends to broader and shift toward higher energy with increasing of nitrogen content DLC. The fitted spectra match very well with the data for DLC films reported previously [48]. The peaks observed at (∼4.4 and 5.1 eV), the first feature is associated with electronic absorption of the sp2 hybridisations, while the second feature, is associated with the combined electronic absorption from both sp2 and sp3 hybrid [50]. As well, the intensity curve of (ε) in N-DLC samples is higher than DLC; this because of the N-DLC surface contains a significant amount of sp2

Table 3 The surface tension components value of film surfaces before and after attachment of glycine. Values before attachment of glycine Samples

 + mJ m−2

DLC NI NII NIII

1.36 1.92 2.15 2.41

␥− mJ m−2 3.17 3.64 3.81 3.86

Values following the exposure of (0.001 M) glycine for 12 h 2.35 15.7 DLC 2.07 15.8 NI 16.2 NII 1.92 2.52 14.4 NIII

(SP )mJ m−2 4.61 5.06 5.17 5.12 11.8 12.2 12.7 10.4

(SD )mJ m−2

 s mJ m−2

Giwi mJ m−2

37.8 44.5 44.9 46.3

42.45 49.5 50.0 51.5

−55.19 −52.46 −50.3 −48. 9

36.6 36.8 36.9 37.1

48.5 49.1 49.6 47.6

−13.24 −13.61 −13.22 −15.51

Key: (Wa ): work of adhesion, ( + ): electron accept (acid component of polar part), ( − ): electron donator (base component of polar part), (SP ): polar part of surface tension, (SD ): dispersive part of surface tension, ( s ): surface free energy), (Giwi ): Gibbs value for hydrophobicity.

5170

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

4

3

DL C NI NII NIII

2

1 2.5

3.0

3.5

4.0

4.5

5.0

5.5

eV Fig. 5. Spectroscopic Ellipsometry of DLC and N-DLC samples.

hybridisation carbon as a compared with DLC film and this value is increased with increased nitrogen content. [51] 5.3. Film characterisation following the exposure of glycine

NH3

NH2

COO

C=O

C-N

2

COO

CH

C-N

+ NH 3 (r)

(str)

(def) C-O C-C

NH

2(t)

(str)

5.3.1. Raman spectroscopy Following the exposure of glycine (0.001 M at pH 7.4), Raman spectroscopy was employed to identify the functional groups i.e. COO− , NH3 + , CH2 , C O, C O, C N, O C O and C C, which are associated with the amino acid Fig. 6 [52]. Bands of the carboxyl group were observed at 588 cm−1 , 835 cm−1 , 1340 cm−1 and 1653 cm−1 [53–55]. Both peaks of C OH and C O are significantly shifted when compared with the normal Raman spectrum of free glycine. The new band observed in both of DLC and N-DLC samples at ∼1727 cm−1 , which was attributed to ester carbonyl group. Asymmetric and symmetric stretching vibration of the COO− in glycine onto DLC, NI and NII samples appeared at ∼1560 cm−1

C-H

2.0

+

1.5

C-H

Dielectric function (ε)

5

and 1387 cm−1 , respectively [56]. Whilst unlikely we didn’t mention this band in case of NIII. The Raman band located at around 1262 cm−1 for glycine on N-DLC can be assigned to the C O and C C symmetric stretching mode [52]. Based on the work of Silva et al., the Raman peak observed at 709 cm−1 in doped and nondoped DLC samples can be attributed to the bending mode of the COO− [57]. The rocking vibration peak of COO appeared at 554 cm−1 [58]. C OH band shifts from 1188 cm−1 (free glycine) to 1203 cm−1 (attached glycine), this could be related to replacement of hydrogen with another atom. The band observed at 430 cm−1 and 491 cm−1 assigned as a torsional mode of NH3 in glycine onto all samples [53], and NH bending vibrations were observed between1609 and 1618 cm−1 [54] Fig. 6. A band appeared 3120 cm−1 and 3220 cm−1 this can be attributed to NH3 + stretching and deprotonated amino group (NH2 ), respectively. The results are agreement with Stewart et al. who studied the interaction of amino acids on the silver surface [55]. The corresponding asymmetric and symmetric stretching bands of the deprotonated amino group (NH2 ) in both of DLC and N-DLC samples have appeared at 3320 cm−1 and 3220 cm−1 , respectively. For DLC, NI and NII samples the strong band at 2943 cm−1 and 2982 cm−1 , can be assigned to symmetric and asymmetric stretching of the protonated amino group (NH3 + ) [56,57,59]. Whilst, in NIII sample these bands were shifted to 2922 cm−1 and 2970 cm−1 , respectively. The results suggest that the chemical reaction has occurred between glycine and surface of samples (DLC and N-DLC) through interaction of carboxyl group to produce glycinate, and the amino group does not play a significant role in the interaction with the surface. Both symmetric and asymmetric (NH3 + ) deformation bands of glycine on DLC and N-DLC samples were lie at 1540 cm−1 (medium) and 1617 cm−1 (strong), respectively [54]. The medium Raman band located at ∼1125 cm−1 and 490 cm−1 is attributed to (N H) rocking frequency and (NH3 + ) torsion mode [59,60]. Basic to the free glycine, (NH3 + ) torsion and rocking vibrations seem to be shifted toward higher wave numbers in both of DLC and NI samples. The effect results from the formation of intermolecular hydrogen bonds (N H· · ·O) in glycine based multi-layers [52,61]. This finding is in agreement with Qu et al., who reported that the breakdown of carboxylic bond (COO H) is thermodynamically

(E )

Intensity (a/u)

(D )

(C )

(B )

(A ) 400

800

1200

1600

2000

2400

2800

3200

Raman Shift cm-1 Fig. 6. Raman spectra of free glycine (A) and attached onto DLC (B), NI (C), NII (D) and NIII (E).

3600

5171

N-H

NH3+

C-H

COOR

+ NH3

COOstr. NH3+ COOR

COOR

CH2

C-N

C-C

C-O(def)

NH2(t)

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

(E)

Absorption

(D)

(C )

(B )

(A) 400

800

1200

1600

2800

3200

3600

Wave number (cm-1) Fig. 7. FTIR spectra of free glycine (A) and attached onto DLC (B), NI (C), NII (D) and NIII (E) film samples.

more stable than breakdown of amino group (HN H) in glycine [62]. The more stable band appeared at ∼1387 cm−1 and ∼1445 cm−1 , in all samples, that corresponded to the symmetric and asymmetric deformation mode of methylene group (CH2 ), respectively [52–54]. As well the CH stretching modes were observed at 2855 cm−1 and 2967 cm−1 . The results correspond to aliphatic compounds containing only one methyl group i.e. glycine [52,60]. Whilst Raman bands arise at 1321 cm−1 and 1154 cm−1 related to deformation and twisting modes of CH2 , respectively [56]. Spectral region (800–1200) cm−1 of glycine onto DLC and N-DLC samples, where backbone C C and C N skeletal stretch coordinates contribute significantly to the normal modes [52,55]. The symmetric stretching vibration of C N is found at 1490 cm−1 and 1017 cm−1 [59,63], and the vibrational band of the primary amine with carbon atom (CH2 NH2 ) has also been identified at 1125 cm−1 [58]. Basic of the band intensity indicated that adsorption of the glycine decreased on samples contains high amount of nitrogen, this is in agreement with other reports published simultaneously [56,57].

Fig. 8. Adsorbed layer of glycine onto surfaces of DLC and N-DLC after 12 h.

5.3.2. FTIR analysis of glycine on DLC Fig. 7 shows the FTIR spectra of adsorbed glycine onto surfaces of DLC and N-DLC samples. The bands at 1557 cm−1 and 1404 cm−1 is slight shifted toward lower wave-number than the expected range due to asymmetric and symmetric vibrations of carboxylate group (COO− ), respectively [64,65]. This shift may be related to the involvement of COO− group to form hydrogen bonds [52]. Band at ∼(1735–1740) cm−1 can be attributed to carboxylate group [65]. A shoulder appears at ∼1700 cm−1 in all samples except (NIII), this is good evidence for interaction between the (COO− ) group of first molecule and the (NH3 + ) group of second molecule of glycine form hydrogen bonds in the second layer of adsorbed glycine onto surfaces, this involving an intermolecular hydrogen bonding and which tend to form glycine multilayer’s onto surfaces. The wave number is shifted somewhat lower, since the carbonyl oxygen is involved in intermolecular hydrogen bonding with one of the hydrogen atoms of the amine moiety [52]. The bending and rocking vibrations of the COO− were observed at ∼705 cm−1 and 519 cm−1 , respectively [57,66]. The band at 1262 cm−1 can be assigned to the C O symmetric stretching mode [67]. The medium intensity bands were observed at 1391 cm−1 and 514 cm−1 , for the samples of DLC, NI and NII, can be assigned to (O C O ) stretch and in-plane of the acetate group, respectively [67,68]. Whilst the band intensity decreases in NIII sample; hence one can conclude that decrease of adsorption ability of glycine onto NIII surface. Strong FTIR bands were located at ∼3400 and ∼3310 cm−1 for DLC and N-DLC samples they were assigned to the asymmetric and symmetric stretching vibrations of deprotonated amine (NH2 ) group [58,67]. The adsorption band at ∼3050 cm−1 in doped and un-doped samples is return to symmetric vibration of (NH3 + ) bond [56,57,59]. Whilst the asymmetric stretching band observed at 3148 cm−1 (NI), 3152 cm−1 (DLC), 3155 cm−1 (NII) and 3160 cm−1 (NIII) [65]. The shift toward lower wave-number might be this related to the amount of protonated form of (NH3 + ). From data above the presence of both protonated and deprotonated amine group onto DLC and N-DLC surfaces, indicated that adsorbed glycine can form a hydrogen bond between both of amine and

5172

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

1.0 0.8

0.5

cos (Δ)

cos (Δ)

0.4

0.0

0.0

-0.4 -0.5 -0.8

DLC Glycine onto DLC 2

3

4

NI Glycine onto NI

-1.0 2

5

3

Energy (eV)

4

5

Energy (eV)

Fig. 9. Spectroscopic Ellipsometry of DLC and NI films before and after attachment of glycine.

5.3.3. Spectroscopic ellipsometry The thickness of adsorbed glycine layer onto samples was obtained, and the values between (17 and 29) A˚ (Fig. 8). The results have shown that the NI sample has a greater layer thickness of glycine, followed by DLC, then NII and NIII. Fig. 9 shows the difference between the samples before and after adsorption of glycine through the relation between cos() vs energy (eV). All of the samples showed that a large change in cos  following the attachment of glycine and this gap increases with increasing thickness of glycine over-layer. There is evidence of a surface morphology change during adsorption of glycine. The results agree with Benesch et al., who reported that a significant decrease in the () value and slight increase of ( ) value, with increasing surface coverage of HSA layers on SiO2 [74]. 5.3.4. Contact angle and surface free energy Followed the adsorption of glycine, the water contact angle values were significant decreased and well dropped to around (60–64) , Fig. 10. These concluded that the surfaces become more hydrophilic, due to the presence of adsorbed glycine. The results from Table 3 show that the surface energy is dramatically increases

Before attachment of glycine After attachment of (0.001M) glycine

80 78

Water contact angle (θ)

carboxyl group in the upper layer. Strong band at ∼1613 cm−1 and 1593 cm−1 confirm the presence of asymmetric (NH3 + ) bending vibration, and a medium band at 1515 cm−1 assigned to symmetric deformation vibration of (NH3 + ) [57,64,69]. A band at ∼1101 cm−1 , can be attributed to the NH3 + rocking vibration, this agrees with theoretical value of (NH3 + ) rocking mode which is located at 1084 cm−1 [66,67]. The shift is return to the influence of H-bond network for glycine [52]. A band at ∼1646 cm−1 is assigned to deformation vibrations of NH2 , [64]. The asymmetric and symmetric stretching vibrations of CH2 groups of DLC and N-DLC samples are located at ∼2993 cm−1 and ∼2930 cm−1 , respectively [54,58,60]. The medium intensity bands at ∼1445, ∼1320 and ∼1159 cm−1 were assigned to scissoring, wagging and twisting deformation frequency of CH2 group, respectively [54,56,59,65]. The stretching bands of C C have been located at 804, 910 and 1369 cm−1 , for both types of DLC and N-DLC samples [53,64,70,71]. The strong and stable bands at ∼880 cm−1 and ∼1075 cm−1 are ascribed to the symmetric and asymmetric stretching vibrations skeletal of C C N, respectively [52,71]. C N stretching modes were observed at around 1511 cm−1 [72]. A band situated at ∼1008 cm−1 can be assigned to C N stretch in addition to C C vibration bond [73].

76 74 72

62 61 60 --

DLC

NI

NII

NIII

Samples Fig. 10. Water contact angle for samples before and after of adsorption of glycine.

followed by adsorption of glycine. As well, the polar component (SAB ) of surfaces is significantly increased, because glycine may exist predominantly as the zwitterions, which tends to form hydrogen bond, in addition of glycine is an ambivalent and it possess a medium hydrophilic [75,76]. These tend to enhance the polar part of surface tension. On the other hand, the results show that the base component values (S− ) are higher than acid component (S+ ), this phenomena is caused by presence of oxygen atoms in glycine molecule. As well, the (Giwi ) value has significantly increased (as compared with samples before attachment of glycine), and the values were between (−13 and −16) mJ m−2 , Table 3. 6. Conclusions Raman analysis of the undoped DLC gave the ratio of (ID/IG) to be 0.58 for DLC. Doping with low levels of nitrogen (5.4 at.%) slightly reduced the ID/IG ratio, while higher levels of nitrogen doping increased the value of this ratio. From FTIR analysis, a band for NC was observed in N-DLC is related to promotion of the sp2 hybridisation content. The surface wettability of the samples showed that DLC possesses slight hydrophobic properties with a water contact angle of ∼79◦ . The water contact angle was markedly decreased following doping with nitrogen. The surface tension of the DLC film was determined to be 42 mJ m−2 , whilst doping with

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

nitrogen tended to enhance of this value to ∼50 mJ m−2 . Spectroscopic ellipsometry analysis has illustrated both (␴) and (␲) bonds in DLC and N-DLC samples. The energy bands of ES in DLC were ∼2.0 eV and ∼3.3 eV, whilst the band gap energy increases in NDLC samples, this can be related to increase of N C and N N bonds in the film structure. A band was observed at 1735 cm−1 in FTIR and at 1727 cm−1 in Raman and these were assigned to the ester bond which proceeds through adsorption of glycine. The FTIR peaks around 3000–3400 cm−1 is attributed to amino bond stretches and the band close to 3000 cm−1 was assigned to NH3 + . Might be these come from the protonated amine of glycine in multilayer form, because in case of monolayer adsorption, following the interaction of the carboxyl group to the surfaces most of amino group is appeared as deprotonated form. Contact angle measurements showed a significant decrease in H2 O ( ◦ ) for samples after adsorption of glycine, this indicates that glycine tends to reduce the hydrophobic property of the DLC films. The Gibbs solid liquid interaction values (Giwi ) were between (−14 and −16) mJ m−2 and is evidence that the surfaces became more neutral, due to glycine possesses a low hydrophobic property in nature. The adsorption of glycine on the surfaces gave an increase in the surface tension, due to hydrogen bonding between the carboxyl and amino functional groups.

Acknowledgement The authors would like to thank all the staffs of Nanotechnology integrated Bio-Engineering Centre (NIBEC) for their assistance with data analysis, and the University of Ulster for funding under their VCRS scheme.

References [1] H.C. Cheng, S.Y. Chiou, C.M. Liue, M.H. Ling, C.C. Chen, K.L. Ou, Journal of Alloys and Compounds 477 (2009) 931–935. [2] T. Hasebe, A. Hotta, H. Kodama, A. Kamijo, K. Takahashi, T. Suzuki, New Diamond and Frontier Carbon Technology 17 (6) (2007) 263–279. [3] A. Alanazi, C. Nojiri, T. Kido, T. Noguchi, Y. Ohgoe, T. Matsuda, K. Hirakuri, A. Funakubo, K. Sakai, Y. Fukui, Artificial Organs 24 (8) (2000) 624–627. [4] M.I. Jones, I.R. McColl, D.M. Grant, K.G. Parker, T.L. Parker, Diamond and Related Materials 8 (1999) 457–462. [5] D. Grimanelis, S. Yang, O. Bohme, E. Roman, A. Alberdi, D.G. Teer, J.M. Albella, Diamond and Related Materials 11 (February (2)) (2002) 176–184. [6] L.J. Yu, X. Wang, X.H. Wang, X.H. Liu, Surface and Coatings Technology 128–129 (1) (2000) 484–488. [7] T. Hasebe, A. Shimada, T. Suzuki, Y. Matsuoka, T. Saito, S. Yohena, A. Kamijo, N. Shiraga, K. Higuchi, H. Yoshimura, S. Kuribayashi, Journal of Biomedical Materials Research 76A (2006) 86–94. [8] H.W. Choi, R.H. Dauskardt, S.C. Lee, K.R. Lee, K.H. Oh, Diamond & Related Materials 17 (2008) 252–257. [9] V.M. Tiainen, Diamond Related and Materials 10 (2001) 153–160. [10] J.I. Onate, M. Comin, I. Braceras, A. Garcia, J.L. Viviente, M. Brizucla, N. Garagorri, J.L. Peris, J.I. Alava, Surface Coatings Technology 142–144 (2001) 1056–1062. [11] R.K. Roy, H.W. Choi, J.W. Yi, M.W. Moon, K.R. Lee, D.K. Han, J.H. Shin, A. Kamijo, T. Hasebe, Acta Biomaterialia 5 (2009) 249–256. [12] S. Lousinian, S. Logothetidis, A. Laskarakis, M. Gioti, Biomolecular Engineering 24 (2007) 107–112. [13] J.W. Ma, J.A. Ruys, S.R. Mason, J.P. Martin, A. Bendavid, Z. Liu, M. Ionescu, H. Zreiqat, Biomaterials 28 (2007) 1620–1628. [14] M. Mohanty, T. Anilkumar, P. Mohanan, C. Muraleedharan, G. Bhuvaneshwar, F. Derangere, Biomolecule Engineering 19 (2002) 125–128. [15] F.Z. Cui, D.J. Li, Surface and Coatings Technology 131 (2000) 481–487. [16] W.J. Ma, A.J. Ruys, R.S. Mason, P.J. Martin, A. Bendavid, Z. Liu, M. Ionescu, H. Zreiqat, DLC coatings: effects of physical and chemical properties on biological response, Biomaterials 28 (2007) 1620–1628. [17] Y. Cheng, Y.F. Zheng, Surface and Coatings Technology 200 (2006) 4543–4548. [18] R.K. Roy, K.R. Lee, Journal of Biomedical Materials Research: Part B: Applied Biomaterials 83B (2007) 72–84, JBMR-B. [19] N. Huang, P. Yang, Y.X. Leng, J. Wang, H. Sun, J.Y. Chen, G.J. Wan, Surface and Coatings Technology 186 (2004) 218–226. [20] A. Hart, E. Schmich, J.A. Garrido, J. Hendrnando, S.C.R. Cathrino, S. Walter, P. Feullner, A. Kromka, D. Steinmuller And, M. Stuitzman, Nature Materials 3 (October) (2004) 736–742. [21] S. Muhl, J.M. Mendez, Diamond and Related Materials 8 (1999) 1809–1830.

5173

[22] C. Wang, S. Yang, J. Zhang, Journal of Non-Crystalline Solids 354 (2008) 1608–1614. [23] A. Erdemir, L. Eryilmaz, M.K. Kazmanli, Surface and Coatings Technology 133–134 (2000) 448–454. [24] A. Grill, Journal of Research and Development 43 (1–2) (1999) 147–161. [25] C.L. Zheng, F.Z. Cui, B. Meng, J. Ge, D.P. Liu, I.S. Lee, Surface and Coatings Technology 193 (2005) 361–365. [26] M. Ahmed, J.A. Byrne, J.A.D. McLaughlin, e-Journal of Surface Science and Nanotechnology 7 (2009) 217–224. [27] X. Wang, L. Yu, C. Li, F. Zhang, Z. Zheng, X. Liu, Colloids and Surfaces B: Biointerfaces 30 (2003) 111–121. [28] E.M. Harnett, J. Alderman, T. Wood, Colloids and surfaces B: Biointerfaces 55 (2007) 90–97. [29] R.A. Silva, M.D. Urzua, F.S. Petri, P.L. Dubin, Langmuir 26 (2010), 14032-14-38. [30] G. Peng, X. Jun, P. Yong, D. Wanyu, D. Xinlu, D. Chuang, Plasma Science and Technology 8 (4) (2006) 425–428. [31] G.T. Gao, P.T. Mikulski, J.A. Harrison, Journal of the American Chemical Society 124 (2002) 7202–7209. [32] R.J. Harmers, Y. Wang, Chemical Reviews 96 (1996) 1261. [33] L. Sun, H.K. Li, G.Q. Lin, C. Dong, Journal of Vacuum Science & Technology A 28 (6) (2010) 1299–1306. [34] C. Wang, S. Yang, J. Zhang, Journal of Nano-Crystalline Solids 354 (2008) 1608–1614. [35] F. Zaera, E.B. Kollin, J.L. Gland, Surface Science 184 (1–2) (1987) 75–89. [36] M.R. Lopez-Ramirez, S. Sanchez-Cortes, M. Perez-Mendez, G. Blanch, Surface Science 41 (10) (2010) 1170–1177. [37] S. Baek, T. Shirafuji, S. Cho, N. Saito, O. Takai, Japanese Journal of Applied Physics 49 (2010) 1–6, 08JF10. [38] H.R. Aryal, S. Adhikari, S. Adhikary, H. Uchida, M. Umeno, Diamond and Related Materials 15 (2006) 1906–1908. [39] X.W. Liu, L.H. Chan, W.J. Hsieh, J.H. Lin, H.C. Shih, Carbon 41 (2003) 1143–1148. [40] W.J. Hsieh, P.S. Shih, C.C. Lin, C.Y. Wang, H.C. Shi, Surface and Coatings Technology 200 (2006) 3175–3178. [41] G.Q. Yu, B.K. Tay, Z. Sun, Surface and Coatings Technology 191 (2005) 236–241. [42] K. Endo, K. Shinoda, T. Tatsumi, Journal of Applied Physics 86 (1999) 2739. [43] C. Liu, Q. Zhao, Y. Liu, S. Wang, E.W. Abel, Colloids and Surfaces B: Biointerfaces 61 (2008) 182–187. [44] R.S. Butter, D.R. Waterman, A.H. Lettington, R.T. Ramos, E.J. Fordham, Thin Solid Films 311 (1997) 107–113. [45] M.C. Howland, A.W. Szmodis, B. Sanii, A.N. Parikh, Biophysical Journal 311 (February) (2007) 1306–1317. [46] M. Vinnichenko, R. Gago, N. Huang, Y.X. Leng, H. Sun, U. Kreissig, M.P. Kulish, M.F. Maitz, Thin Solid Films 455–456 (2004) 530–534. [47] H. Forster, Mol Sieves, vol. 4, Springer-Verlag, Berlin Heidelberg, 2004, pp. 337–426. [48] H. Arwin, Thin Solid Films 377–378 (2000) 48–56. [49] G. Fedosenko, D. Korzec, J. Engemann, D. Lyebyedyev, H.-C. Scheer, Thin Solid Films 406 (2002) 275–281. [50] S. Logothetidis, M. Gioti, S. Lousinian, S. Fotiadou, Thin Solid Films 482 (2005) 126–132. [51] N. Menegazzo, M. Kahn, R. Berghauser, W. Waldhauser, B. Mizaikoff, Analyst 136 (2011) 1831–1839. [52] S. Pandiarajan, M. Umadevi, R.K. Rajaram, V. Ramakrishnan, Spectrochimica Acta Part A 62 (2005) 630–636. [53] D. Sajan, J. Binoy, B. Pradeep, K.V. Krishna, V.B. Kartha, I.H. Joe, V.S. Jayakumar, Spectrochimica Acta Part A 60 (2004) 173–180. [54] T.V. Kumar, I.H. Joe, C.P.R. Nair, V.S. Jayakumar, Journal of Molecular Structure 877 (2008) 20–35. [55] S. Stewart, P.M. Fredericks, Spectrochimica Acta Part A 55 (1999) 1641–1660. [56] J. Baran, H. Ratajczak, Vibrational Spectroscopy 43 (2007) 125–139. [57] B.L. Silva, P.T.C. Freire, F.E.A. Melo, I. Guedes, M.A.A. Silva, J.M. Filho, A.J.D. Moreno, Brazilian Journal of Physics 28 (1) (1998) 19–24. [58] A.E. Aliaga, C. Garrido, P. Leyton, G. Diaz, F.J.S. Gomez-Jeria, T. Aguayo, E. Clavijo, M.M.C. Vallette, S.S. Cortes, Spectrochimica Acta Part A 76 (2010) 458–463. [59] S. Ramaswamy, R.K. Rajaram, V. Ramakrishnan, Journal of Raman Spectroscopy 34 (2003) 50–56. [60] S.V. Goryainova, E.N. Kolesnika, E.V. Boldyreva, Physica B 357 (2005) 340–347. [61] A. Pawlukojc, J. Leciejewicz, J. Tomkinson, S.F. Parker, Spectrochimica Acta Part A 57 (2001) 2513–2523. [62] Y.Q. Qu, Y. Wang, J. Li, K.L. Han, Surface Science 569 (2004) 12–22. [63] E. Ramachandran, S. Natarajan, Crystal Research and Technology 42 (6) (2007) 617–620. [64] D. Sajan, N. Vijayan, K. Safakath, R. Philip, I.H. Joe, Journal of Physics Chemistry A 115 (2011) 8216–8226. [65] A. Huczynski, M. Ratajczak-Sitarz, J. Stefanska, A. Katrusiak, B. Brzezinski, F. Bartl, The Journal of Antibiotics 64 (2011) 249–256. [66] G.S. Nichol, J.H. Paredes, H.E.E. Poncce, M.P. Beltran, M.E.A. Ramos, A.D. Molle, Revista Mexicana De Fisica S 54 (1) (2008) 13–16. [67] G. Tzvetkov, G. Koller, Y. Zubavichus, O. Fuchs, M.B. Casu, C. Heske, E. Umbach, M. Grunze, M.G. Ramsey, F.P. Netzer, Langmuir 20 (2004) 10551–10559. [68] D. Wang, S.B. Lei, L.J. Wan, C. Wang, C.L. Bai, Journal of Physics Chemistry B 107 (2003) 8474–8478. [69] E. Gallegos-Loya, A. Duarte-Moller, International Journal of the Physical Sciences 6 (14) (2011) 3310–3318. [70] P. Taddei, F. Balducci, R. Simoni, P. Monti, Journal of Molecular Structure 744 (2005) 507–514.

5174

M.H. Ahmed, J.A. Byrne / Applied Surface Science 258 (2012) 5166–5174

[71] M.C. Corbeil, A.L. Beauchamp, S. Alex, R. Savoie, Canadian Journal Chemistry 64 (1986) 1876–1884. [72] A.G. Zavaglia, R. Fausto, Physical Chemistry Chemical Physic 5 (2003) 3154–3161. [73] M. Ibrahim, A. Nada, D.E. Kamal, Indian Journal of Pure and Applied Physics 43 (2005) 911–917.

[74] J. Benesch, A. Askendal, P. Tengvall, Journal of Colloid and Interface Science 249 (2002) 84–90. [75] J.A. Lori, T. Hanawa, Corrosion Science 43 (November (11)) (2001) 2111–2120. [76] B.W. Matthews, Encyclopaedia of Life Sciences (April) (2001) 1–6.