Gold nanoparticles immobilization: Evidence of amination of diamond surfaces in liquid ammonia

Diamond & Related Materials 32 (2013) 36–42

Contents lists available at SciVerse ScienceDirect

Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond

Gold nanoparticles immobilization: Evidence of amination of diamond surfaces in liquid ammonia G. Charrier, D. Aureau, A.-M. Gonçalves, G. Collet, M. Bouttemy, A. Etcheberry, N. Simon ⁎ Institut Lavoisier de Versailles, UMR 8180, Université de Versailles Saint-Quentin en Yvelines, 45 avenue des Etats-Unis, 78000 Versailles, France

a r t i c l e

i n f o

Article history: Received 6 July 2012 Received in revised form 15 November 2012 Accepted 30 November 2012 Available online 7 December 2012 Keywords: Diamond films BDD Amination Liquid ammonia Amino groups Oxygen terminations XPS Gold nanoparticles

a b s t r a c t Diamond presents many interesting properties in terms of biocompatibility, chemical inertness and is attractive for applications notably in the biosensor field. The formation of amine functional groups at diamond surface is of great interest for the incorporation of biomolecules. In the present work, we present an efficient direct amination method using anodic treatment in liquid ammonia on oxygen terminated diamond surface. Chemical surface modifications were followed by XPS analyses, while the distribution of amino-groups was indirectly studied by gold nanoparticles immobilization. The resulting surfaces were examined using FEG-SEM imaging. This combined approach has been powerful evidencing both the presence of notable amounts of amino groups after the electrochemical treatment in NH3liq and their homogeneous distribution at the diamond surface. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The detection of biomolecules requires materials with specific properties as biocompatibility, stability and sensitivity [1]. Semi-conductive diamond is highly biocompatible, chemically inert and presents a very large electrochemical window, making it an attractive substrate for biosensing [2–9]. As described in the literature, amine terminal groups would be of great interest for the immobilization of biomolecules [10,11]. The surface termination is one of the most important parameters affecting the properties of diamond [12–14]. Several accurate methods relating to the modification of as-grown H-terminated boron-doped diamond (BDD) are described in the literature. Even if as-deposited hydrogen terminations are rather stable, they can be converted to nitrogen terminations, either by dry or wet techniques. For example, interesting preconditioning of the diamond surface can be reached on H-BDD with the diazonium spontaneous reaction [15,16]. Several amination methods of diamond films are thus proposed in the literature. Two main strategies have been developed. On one hand, the covalent bonding of amine terminated alkyl chains has been investigated using for example silanes [17–23] or diazonium salts [14,24–29]. These methods can be seen as indirect ones because they provide an organic spacer between the surface and the functional group. On the other hand, more direct methods, where the nitrogencontaining group is directly bonded to diamond have been developed, ⁎ Corresponding author. Tel.: +33 139254403. E-mail address: [email protected] (N. Simon). 0925-9635/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.diamond.2012.11.014

as for example the chemical treatment of a chlorinated surface with ammonia gas [30,31], the UV irradiation in ammonia gas [32] or the use of radiofrequency plasmas of mixtures He+NH3 [33–35]. In 2009, our team reported the formation of nitrogen containing groups at diamond surface after an anodic polarization in liquid ammonia [36]. As small N concentrations were detected after this process, the increase of the direct amination efficiency in NH3liq was still a challenge. Recently, works evidenced that the presence of oxygen either in the gas phase or bonded to the diamond surface was essential to obtain surface amination in one step UV \NH3 gas process [37]. Based on these results, we propose to study in the present work the direct amination of diamond surface by anodic treatments in liquid ammonia on both as-grown H-terminated and previously oxidized surfaces. The formation of surface nitrogen-containing groups was followed by XPS measurements. As a homogeneous surface distribution of amino groups is required for functionalization purposes, the repartition of these functional groups has been investigated by using an indirect probe: the deposition of gold nanoparticles. The resulting samples were examined by scanning electron microscopy and XPS analysis. 2. Experimental 2.1. Preparation of the diamond surfaces The polycrystalline boron-doped diamond (BDD) films, deposited on silicon substrates in a hot filament-assisted chemical vapor deposition process, were provided by Adamant (Neuchatel, Switzerland).

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

2.2. Preparation and deposition of the gold nanoparticles Gold nanoparticles (AuNPs) were synthesized using the TurkevichFrens [40,41] method under conditions yielding 15 nm nanoparticles stabilized by a citrate layer in aqueous solution. A 1 mL solution of citrates (8.5×10−4 mol L−1) was added to a 19 mL solution of HAuCl4 (2.5×10−4 mol L−1) at 80 °C under vigorous stirring. The stirring was maintained for 45 min at 80 °C (color gradually changing from slight yellow to gray, purple, and finally dark red). Then, the solution was cooled down to room temperature and finally sonicated. This well-documented method yields reproducible monodispersed AuNPs with average diameter of 15 nm (standard deviation less than 1 nm) [42,43]. The aggregation of AuNPs is prevented by the negatively charged citrate layer surrounding the particles. However, citrates do not form a covalent bond with gold and can be easily replaced by other molecules or chemical groups anchored on surfaces. AuNPs were subsequently deposited from the solution to the sample by dipping the diamond surfaces (both H- and O-BDD) into the gold colloidal solution for 90 min. Under these conditions, the pH of the solution is around 5.5, so that the amino-terminated surfaces (pKa≈10) are protonated (acid form: NH3+) in contact with the gold suspension [42,43]. The quality (size, shape, and dispersion) of the nanoparticles solution is characterized by UV–visible absorption (see Fig. 1). When the surface plasmon resonance (SPR) is monitored, which is dependent on these properties [44], a systematic control of the particle size and dispersion is possible. For 15 nm particles, for instance, the SPR has a sharp maximum around 520 nm. Importantly, to ensure that the AuNPs are firmly attached and to avoid multilayers, the surface is sonicated during 15 min after deposition. 2.3. Instrumentation XPS measurements were performed on a VG220i XL system with a base pressure of 5×10−10 Torr and using the AlKα (1486.5 eV) X-ray monochromatized radiation with a pass energy of 20 eV or 8 eV (resolution 0.2 eV). The incidence angle of the X-ray probe was 90°. Energy levels of XPS were calibrated with Au single crystal. The spectra were processed using the VG Eclipse Data system.

0,8 0,7

520 nm 0,6

Absorbance

The morphology of these surfaces was studied by AFM and XPS. Grains are around 1 μm large. The surface roughness is about hundreds of nanometers. The number of boron atoms in the diamond layer was determined to be 1× 1020 B cm −3 by SIMS measurements. Because of the deposition process, the surface is mainly hydrogenated and samples will be termed H-BDD. Samples were oxidized by immersion into an oxidizing solution: MnO4− 0.1 M in H2SO4 0.5 M [38]. Potassium permanganate (KMnO4) was purchased from Aldrich and used without further purification. Sulfuric acid (H2SO4) was provided by Prolabo (Normapur quality). Oxidized surfaces will be named O-BDD. Amination of H-BDD and O-BDD surfaces was obtained by an electrochemical anodization in liquid ammonia. A galvanostatic treatment with a current density Ja = 10 mA cm−2 was applied during 500 s. Ammonia condensation, from gaseous ammonia (“electronic grade” from Air Liquide) was provided by a glass column assembly and required a low operating temperature under atmospheric pressure. An electrochemical cell filed up with 150 cm3 of liquid ammonia was maintained at 213 K in a cryostat. The acidic medium was obtained by addition of 0.1 M NH4Br (purest available quality from Aldrich). All potentials were measured vs. a silver reference electrode [39]. After the treatment, electrodes were rinsed in the purest liquid ammonia and then in milliQ water (18 MΩ cm). The samples were finally handled under a water drop protection before being dried under a nitrogen gas stream and being transferred towards an XPS analyzer using a procedure avoiding any air contamination. Aminated surfaces will be termed N-BDD.

37

0,5 0,4 0,3 0,2 0,1 0,0 400

500

600

700

800

Wavelength (nm) Fig. 1. UV spectroscopy of gold nanoparticle solution.

SEM micrographs were acquired with a Jeol JSM 7001-F at 5 kV accelerating voltage. 3. Results and discussion 3.1. Amination treatment in liquid ammonia In a preliminary work [36], we showed that it was possible to create carbon–nitrogen bonds on diamond films by anodic treatments in liquid ammonia. Small but encouraging amounts of nitrogen were detected by XPS measurements (about 3 XPS % atomic) after anodization in NH3liq of as grown H-BDD samples [36]. Besides, recent work [37], suggests that amination treatments of diamond surfaces are more efficient in the presence of oxygen. So, in order to optimize the direct amination of diamond film in liquid ammonia, electrochemical treatments have been studied on both previously oxidized BDD samples (O-BDD) and as-grown H-BDD for comparison. The resulting surfaces will be now respectively named: N + O-BDD and N + H-BDD. Therefore, some as-grown samples were first oxidized using a process already described in a previous work [38]. H-BDD electrodes were immersed into a solution containing 0.1 M of MnO4− (as potassium) in 0.5 M H2SO4, for 2 days. The resulting O-BDD surfaces were then exposed to the anodic treatment in liquid ammonia. To get qualitative and quantitative information about the resulting Nitrogen functionalities after amination treatment, XPS analyses were performed on BDD interfaces before and after the different steps: oxidation and amination. Various current densities ranging from 1 μA cm−2 to 100 mA cm−2 and durations (1 to 30 min) have been tested. When Ja b 1 mA cm−2, whatever the duration (up to 30 min) XPS analysis does not reveal the formation of nitrogen containing groups. At the opposite when Ja becomes too high (100 mA cm−2), the result is the same, the XPS signal of N is absent. Finally the best results are obtained for an intermediate current density of 10 mA cm−2 and duration of 500 s. Similar observations have been made concerning the effect of current density on the formation of C\O groups after anodic processes in aqueous media [45], best coverage of BDD surfaces with oxygenated groups were obtained for electrodes oxidized with intermediate current densities. Both chronopotentiograms corresponding to the electrochemical surface activation with Ja =10 mA cm−2 on H- and O-BDD in liquid ammonia are presented in Fig. 2.On both graphs, the voltage strongly increases from the open circuit potential values (respectively ~0.9 V/SER and ~1.2 V/SER for H-BDD and O-BDD) up to a similar value ~2.5 V/SER for the two kinds of samples. The potential reached during the first seconds of treatment remains quite stable during the whole process. The evolution of BDD surfaces submitted to these processes will be described in the following parts of the manuscript. Figs. 3 and 4 show the surveys obtained respectively for as-grown and oxidized diamond surfaces before and after the amination process.

38

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

2,8

2,6

2,6

2,2

2,4

2,0

2,2

E(V/SRE)

E (V/SRE)

2,4

1,8 1,6 1,4 1,2

2,0 1,8 1,6 1,4

1,0

Treatment on H-BDD

0,8

1,2

Treatment on O-BDD

1,0 0

100

200

300

400

500

0

100

t(s)

200

300

400

500

t(s)

Fig. 2. E = f(t) graphs registered during the anodic treatments on H-BDD (left) and O-BDD (right).

The total carbon, oxygen and nitrogen amounts were obtained by dividing the relevant peak integrated areas of well defined C1s, O1s and N1s core level spectra by the appropriate bulk sensitivity factors following the method described in Ref. [46]. Then, the relative atomic concentrations of carbon, oxygen and nitrogen on the surface were calculated for each BDD sample (see Table 1). After the oxidation process, the oxygen amount increases from 5 XPS % atomic on an as-grown sample to 18%. As already described in Ref. [38], XPS measurements performed on these O-BDD surfaces evidence that mostly singly bounded C\O functions are present, i.e. ether or hydroxyl groups. Nitrogen remains hardly detectable on both as grown and oxidized diamond films. After the electrochemical treatment in liquid ammonia, a new contribution appears around 400 eV attributed to N1s peak (see Figs. 2, 3 and 4) for both N+H-BDD and N+O-BDD. Quantitative analysis shows that the XPS atomic % of nitrogen reaches 5% and 7% after the process in NH3liq performed respectively on H-BDD and O-BDD. This result confirms that the presence of oxygen seems to have an influence on the amination process as already described [37]. In addition, it should be emphasized that the nitrogen concentration detected here on N+O-BDD is similar to the maximum nitrogen concentrations reported for other direct amination processes. Indeed, to our knowledge, Denisenko et al. [35] reported on monocrystalline diamond the most important nitrogen concentration obtained for direct amination with a ratio [N]/[C]=0.063 detected by XPS. They estimate after simulation that it corresponds to a surface density of nitrogen atoms of 2.4·1015 cm−2 which is very close to the density of carbon atoms at (100) surface meaning coverage by a full monolayer. Even if the XPS atomic ratios detected on our polycrystalline samples can't be exactly compared to results obtained by Denisenko et al. on monocrystalline samples, the XPS analysis conditions being the same (incidence angle of the X-ray probe of 90°) the nitrogen atomic concentration measured after electrochemical treatment in liquid ammonia

(7%) is a very encouraging value meaning that the coverage of BDD surface with C\N groups is probably very high. For control experiments, XPS measurements also have been performed on H-BDD and O-BDD samples simply exposed to liquid ammonia for 30 min. The results show no differences compared to that obtained on the H- or O-BDD surfaces non exposed to NH3liq; the [N] remains always below 0.5 XPS at.%. These control experiments show that there is no specific adsorption of ammonia molecules on H- or O-BDD surfaces, evidencing the crucial role of anodic process in the evolution of the nitrogen concentration at BDD interfaces. Concerning the quantity of oxygen (Table 1), a quasi constant value is detected by XPS after each amination process, suggesting that the nonintentional oxidation of the surface which could occur when the sample is extracted from the electrochemical cell is minor. Fig. 5 shows a zoom on the N1s region of the XPS spectrum. The associated fit clearly establishes that two main components with the same full width at half maximum (FWHM) are detected at 399 eV (N1) and 401 eV (N2) on both types of surface (N+H- and N+O-BDD). According to the literature, peak named N1, situated at 399 eV, can be ascribed to “C\NH2” terminations [47–50]. Such chemical groups are especially interesting for biofunctionalization due to the accessibility of the N\H bonds. The attribution of the N2 peak at 401 eV is less evident. Such binding energy can be assigned to nitrogen atoms linked to 2 carbon atoms, as double carbon–nitrogen bonds “C_N” or bridges “C\N\C” [33]. Concerning the peak intensities of the two components N1 and N2, their proportion seems not dependent on the duration of the electrochemical activation nor on the surface termination of the starting BDD interfaces (either H\or O\). Their proportion varies from 40 to 60% for each peak from an experiment to another on a sporadic way.

1200 1200

1000

800

600

400

200

0

1000

800

600

400

200

0

Binding energy (eV)

Binding energy (eV) Fig. 3. XPS surveys of an as-grown (black) and an aminated (light gray) diamond surface.

Fig. 4. XPS surveys of an oxidized (black) and an oxidated + aminated (gray) diamond surface.

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

probe for the detection of amine groups. AuNPs were synthesized with the Turkevich method, yielding 15 nm particles stabilized by a citrate layer. The negatively charged citrates play two roles: they ensure repulsion between the particles to avoid aggregation and they produce an attractive force with the protonated “C\NH3+” at the surface of the diamond film. Citrates do not form a covalent bond with gold; their labile character allows the formation of a direct bond between gold and the amino-terminated surface. The important rinsing, used after the deposition process, ensures that the detected nanoparticles are really attached on the surface and not only physisorbed. Four different types of diamond samples were immersed into the colloidal solution for 1 h30: as-grown H-BDD, oxidized O-BDD, N+H-BDD and N+O-BDD. The resulting surfaces were analyzed by XPS and imaged by SEM. As already reported, as-grown surfaces are mainly H-terminated, with a very small amount of nitrogen (b0.5 XPS at.%). After the AuNPs deposition process on this surface, XPS analysis does not exhibit any gold peak. Fig. 5 shows a SEM image of this H-BDD sample. No gold nanoparticles were observed on this image and on the several other spots probed on the whole surface. Similar results were obtained with the oxidized O-BDD sample. After the oxidation treatment, about 18% of oxygen is detected by XPS measurements but nitrogen remains in trace amount. After the AuNPs deposition process, no gold is detected by XPS and no gold nanoparticles are observed by SEM imaging (Fig. 6). These results confirm that gold nanoparticles do not bind to “C\H” or “C\O” groups. N + H-BDD and N + O-BDD samples, with respectively 5 and 7% N at the surface, were exposed to the gold nanoparticles solution for the same duration. XPS analyses show that the nitrogen peaks are still detected. In addition, peaks associated with gold atoms appear on both kinds of surfaces. Fig. 7 shows a zoom on the Au4f region of the

Table 1 XPS atomic % for an as-grown (H-BDD), an oxidized (O-BDD), an aminated (N+H-BDD) and an oxidated+aminated (N+O-BDD) diamond surface.

C (XPS atomic %) O (XPS atomic %) N (XPS atomic %)

H-BDD

O-BDD

N + H-BDD

N + O-BDD

95 5 b0.5

82 18 b0.5

88 7 5

77 16 7

The presence of the N1 peak attributed to the formation of “C\NH2” terminations at the diamond surface is a really encouraging result for post-functionalization purposes. In order to optimize the binding of biomolecules on this surface, the remaining question is the repartition of the functional groups. Indeed, a homogeneous coverage is required for the attachment of biomolecules, especially on polycrystalline materials where grain boundaries are not easily accessible to large molecules. If reactive “C\NH2” groups are spread on the whole surface, the linking of molecules will be favored. The distribution of the amine groups on the surface was studied using an indirect visual probe: the gold nanoparticles deposition. 3.2. Immobilization of gold nanoparticles: distribution of amino groups Several published works show the specific interaction between gold and amine groups [51–54]. The method recently developed on monolayers on silicon appears efficient to characterize the density of amine groups on surfaces [53]. The nature of the bond between AuNP and the amino group is not clearly established, although it probably involves the lone pair of the nitrogen atom [51–55]. Thus, the immobilization of gold nanoparticles at the diamond surface appears to be a good indirect

3000

N+H-BDD

N+O-BDD

3400

2800

3200

Counts/s

Counts/s

39

2600

3000 2800

N2

2400

N2

N1

N1

2600

2200 2400 404

402

400

398

396

404

Binding energy (eV)

402

400

398

396

Binding Energy (eV)

Fig. 5. N1s XPS spectrum of an aminated (left) and an oxidated + aminated diamond surface (right).

H-BDD

O-BDD

Fig. 6. SEM images of an as-grown H-BDD sample (left) and an oxidized O-BDD sample (right) submitted to the gold nanoparticles deposition process.

40

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

400

Au 4f N+O-BDD after AuNP

240

Au 4f N+H-BDD after AuNP Counts/s

Counts/s

200

200

160 120 80

0 40 92

90

88

86

84

82

80

92

90

Binding energy(eV)

88

86

84

82

80

Binding energy (eV)

Fig. 7. Au4f XPS spectrum of an aminated (left) and an oxidated + aminated (right) diamond surface submitted to the gold nanoparticles deposition process.

XPS spectrum. The proportion of the different chemical species C, O, N and Au is given in Table 2. The resulting surfaces were imaged by SEM (Figs. 8 and 9), evidencing the presence of gold nanoparticles successfully attached on both amino-terminated surfaces. A more important proportion of nanoparticles are clearly visible on N + O-BDD compared to N + H-BDD in agreement with the nitrogen concentrations detected on both surfaces. In both cases, they are firmly immobilized as sonication does not remove them and remain stable in air for months. They are spread on the entire surface. There is no preferential attachment in grain boundaries. The homogeneity of the covering was checked by probing numerous spots on the surface, revealing a uniform repartition of the particles all over the diamond film. SEM images confirm that the AuNPs have a diameter of approximately 15 nm, in agreement with the size determined by UV spectroscopy after the synthesis (Fig. 1). Such observation is important in order to control that the nanoparticles have not been modified during the deposition process. Isolated particles are largely predominant, confirming that no gold clusters have been formed by precipitation, suggesting that the citrate layer is still surrounding the NPs. The surface AuNPs density was calculated from SEM pictures. The samples exhibit a quite reproducible density with values about 109 and 3×109 AuNP cm−2 respectively for N+H×-BDD and N+O-BDD. The average distance between deposited particles is around 300– 800 nm and 150 nm respectively for N + H-BDD and N + O-BDD, i.e., the surface is far from being completely covered in both cases. The electrostatic repulsion between the negatively charged citrate molecules surrounding the particles cannot be the only explanation for such a scattered distribution. The amount of gold detected from XPS measurements is very low (0.3–0.5 XPS at.%). Besides, there is still a notable amount of nitrogen (about 3 and 4.5 XPS at.%) after the deposition process. The use of other deposition methods, as the drop meniscus has been tested without improvement of the deposition. It could thus be considered that the concentration and the deposition method are not the limiting reason for such low densities because in ref [52], really higher nanoparticle surface densities were obtained on surfaces densely covered by NH2 groups starting with solution of the exact same concentration. Only a part of the nitrogen containing groups formed at the diamond surface during the treatment in liquid ammonia are thus in interaction Table 2 XPS atomic % for an aminated and an oxidated+aminated diamond surface after the gold nanoparticles deposition process.

C (XPS atomic %) O (XPS atomic %) N (XPS atomic %) Au (XPS atomic %)

N + H-BDD after Au-NPs deposition

N + O-BDD after Au-NPs deposition

90 7 3 0.2

79 16 4.5 0.5

with a Au-NP. One explanation could be that due to the size of Au-NPs employed in this work (diameter ~15 nm) more than one NH2 function is necessary to immobilize one NPs-Au. Several amine groups favorably disposed should then be required for the immobilization of just one gold nanoparticle; consequently this attachment cannot occur if the “C\NH2” groups are too distant. Finally, the calculation of the density of amino-groups will be very interesting, but cannot be deduced from the density of Au-NPs. The study of the immobilization of gold-NPs with much smaller particles (∅ ~ 2 nm) would be very interesting if we can admit that one C\NH2 would be required to immobilize one Au-NP. But the problem with small NPs is that the surfactant agent used to stabilize them is generally very stable all around the NPs [56] and thus it does not move to allow a direct interaction between Au and NH2. In our work the citrate surfactant is very labile and allows this interaction. The method remains, however, a good way of evidencing the presence and the distribution of amino-groups on the surface. After the anodic treatment in liquid ammonia, numerous amino groups then are formed. They are well-distributed and spread on the whole surface, which is a real asset for the post-functionalization of the surface with more complex structures such as biomolecules. 3.3. Mechanism Even if the efficiency is different, this work shows that amination occurs both on O-BDD and H-BDD. Additionally, the fact that the oxygen atomic concentration remains constant after the growth of amino groups whereas the carbon atomic fraction decreases (see Table 1) gives information on the way amine radicals should react at BDD surface. All these results suggest that the formation of C\NH2 groups takes place from an interaction between C\H groups and amino radicals produced during the oxidation process (Eq. (1)). Very similarly to a mechanism proposed to explain the formation of hydroxyl groups during an anodic process in aqueous media [57], the formed NH2• can abstract hydrogen from the diamond interface forming ammonia and a radical site on the diamond, which will abstract another NH2• (Eq. (2)). NH3 þ h

þ

BV →NH2 ·

þH

þ

diamond–H þ NH2 · →diamond–· þ NH3 diamond–· þ NH2 ·→diamond–NH2 :

ð1Þ ð2Þ ð2Þ

Even if the formation of amino groups seems to occur from C\H terminations, our work shows the beneficial influence of the presence of oxygen terminations at BDD surfaces. As already published by Torrengo et al. [37] the presence of a high electronegative element seems to favor the amination mechanism, but the way it acts is not yet elucidated.

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

41

Fig. 8. SEM images of an aminated sample after immersion into a gold nanoparticles solution.

Fig. 9. SEM images of an oxidated + aminated sample after immersion into a gold nanoparticles solution.

4. Conclusion The direct amination of diamond surfaces through anodic treatment in liquid ammonia has been investigated on both hydrogen and oxygen terminated diamond surfaces. The chemical surface modifications were followed by XPS analyses, while the distribution of amino-groups was indirectly studied by gold nanoparticles immobilization. The XPS measurements reveal the presence of notable nitrogen concentrations on both kinds of surfaces (H- and O-BDD). The highest Nitrogen concentration (7 XPS at.% ) is measured for the O-BDD surface, reaching a value close to the maximum obtained with other direct amination methods which proves that electrochemical treatments in NH3liq are an efficient way to produce “C\N” functional groups at diamond surfaces. After the process in NH3liq, two contributions are visible on the N1s spectra of the different samples, suggesting the presence of different “C\N” functionalities, either amino, imino or amido groups.... As a specific interaction exists between gold and C\NH2 groups, the study of the immobilization of gold nanoparticles on both kinds of BDD surfaces anodically treated in liquid ammonia (N + H- and N + O-BDD) was used to specifically probe the presence of amino groups. The presence of numerous and well dispersed Au-NPs has been detected by SEM-FEG for all the BDD surfaces with a largest proportion for the surface previously oxidized, confirming the formation of C\NH2 groups and their homogeneous distribution at BDD surfaces which is of great interest for functionalization purposes. From the different results, a mechanism for the formation of amino groups has been proposed based on an interaction between C\H groups and amino radicals produced during the oxidation process. As already

published by Torrengo et al. [37] the presence of a high electronegative element seems to favor the amination mechanism, but the way it acts is not yet elucidated. To conclude, the combined approach of XPS measurements and immobilization of Au-NPs used in this work has been powerful evidencing the formation by anodic treatments in liquid ammonia of well-dispersed amino groups, with notable proportions, especially when the BDD surface has been previously oxidized. Prime novelty In this paper, we present an efficient direct amination method using anodic treatment in liquid ammonia on oxygen terminated diamond surface. Chemical surface modifications were followed by XPS analyses, while the distribution of amino-groups was indirectly studied by gold nanoparticles immobilization. The resulting surfaces were examined using FEG-SEM imaging. This combined approach has been powerful evidencing both the presence of notable amounts of amino groups after the electrochemical treatment in NH3liq and their homogeneous distribution at the diamond surface. References [1] Y.V. Pleskov, A.Y. Sakharova, et al., J. Electroanal. Chem. 228 (1987) 19–27. [2] T. Strother, T. Knickerbocker, J.N. Russel, J.E. Butler, L.M. Smith, R.J. Hamers, Langmuir 18 (2002) 968–970. [3] W. Yang, O. Auciello, J.E. Butler, W. Cai, J.A. Carlisle, J.E. Gerbi, D.M. Gruen, T. Knickerbocker, T.L. Lasseter, J.N. Russel, L.M. Smith, R.J. Hamers, Nat. Mater. 1 (2002) 253–257. [4] T. Knickerbocker, T. Strother, M.P. Schwartz, J.N. Russel, J. Butler, L.M. Smith, R.J. Hamers, Langmuir 19 (2003) 1938–1942.

42

G. Charrier et al. / Diamond & Related Materials 32 (2013) 36–42

[5] B. Sun, S.E. Baker, J.E. Butler, H. Kim, J.N. Russel, L. Shang, K.Y. Tse, W. Yang, R.J. Hamers, Diamond Relat. Mater. 16 (2007) 1608–1615. [6] K. Ushizawa, Y. Sato, T. Mitsumori, T. Machinami, T. Ueda, T. Ando, Chem. Phys. Lett. 35 (2002) 105–108. [7] N. Yang, H. Uetsuka, J. Yu, E. Osawa, N. Tokuda, C.E. Nebel, Diamond Relat. Mater. 19 (2009) 910–917. [8] JianWen Zhao, JinLiang Wang, JinFang Zhi, ZhiMing Zhang, Sci. Chin. Chem. 53 (2009) 1378–1384. [9] G.J. Zhang, K.S. Song, Y. Nakamura, T. Ueno, T. Funatsu, I. Ohdmari, H. Kawarada, Langmuir 22 (2006) 3728–3734. [10] G.J. Zhang, K.S. Song, et al., Langmuir 22 (2006) 3728–3734. [11] J. Wang, J.A. Carlisle, Diamond Relat. Mater. 15 (2005) 279–284. [12] I. Yagi, H. Notsu, T. Kondo, D.A. Tryk, A. Fujishima, J. Electroanal. Chem. 473 (1999) 173. [13] K. Hayashi, S. Yamanaka, H. Watanabe, T. Sekiguchi, H. Okushi, K. Kajimura, J. Appl. Phys. 81 (1997) 744. [14] J.B. Cui, J. Ristein, L. Ley, Appl. Phys. Lett. 81 (1998) 429. [15] S.Q. Lud, M. Steenackers, R. Jordan, B. Paola, P. Feulner, J.A. Garrido, M. Stutzmann, J. Am. Chem. Soc. 128 (2006) 16884–16891. [16] A. Bongrain, C. Agnès, L. Rousseau, J.C. Arnault, S. Ruffinatto, F. Omnes, P. Mailley, G. Lissorgues, P. Bergonzo, Langmuir 27 (2011) 12226–12234. [17] H. Notsu, T. Fukazawa, T. Tasuma, D.A. Tryck, A. Fujishima, Electrochem. Solid State 4 (2001) H1. [18] Y. Coffinier, S. Szunerits, B. Marcus, R. Desmet, O. Melnyk, L. Gengembre, E. Payen, D. Delbouglise, R. Boukherroub, Diamond Relat. Mater. 16 (2007) 892–898. [19] G. Zhang, H. Umezawa, H. Hata, T. Zako, T. Funatsu, I. Ohdomari, H. Kawarada, Jpn. J. Appl. Phys. 44 (2005) L295–L298. [20] S.E. Pust, S. Szunerits, R. Boukherroub, G. Wittstock, Nanotechnology 20 (2009) 075302. [21] J.H. Moon, J.W. Shin, S.Y. Kim, J.W. Park, Langmuir 12 (1996) 4621–4624. [22] A. Pallandre, K. Glinel, A.M. Jonas, B. Nysten, Nano Lett. 4 (2004) 365–371. [23] J. Hernando, T. Pourrostami, J.A. Garrido, O.A. Williams, D.M. Gruen, A. Kromka, D. Steinmuller, M. Stutzmann, Diamond Relat. Mater. 16 (2007) 138–143. [24] T.C. Kuo, R.L. McCreery, G.M. Swain, Electrochem. Solid State 2 (1999) 288–290. [25] W. Yang, S.E. Baker, J.E. Butler, Chem. Mater. 17 (2005) 938–940. [26] J. Wang, M.A. Firestone, O. Auciello, J.A. Carlisle, Langmuir 20 (2004) 11450–11456. [27] D. Shin, N. Tokuda, B. Rezek, C.E. Nebel, Electrochem. Commun. 8 (2006) 844–850. [28] G. Shul, P. Actis, B. Marcus, M. Opallo, R. Boukherroub, S. Szunerits, Diamond Relat. Mater. 17 (2007) 1394–1398. [29] S. Szunerits, R. Boukherroub, J. Solid State Electrochem. 2 (2008) 1205–1218. [30] J.B. Miller, D.W. Brown, Langmuir 12 (1996) 5809–5817. [31] T. Ando, R.E. Rawles, K. Yamamoto, M. Kamo, Y. Sato, M. Nishitani-Gamo, Diamond Relat. Mater. 5 (1996) 1136–1142. [32] G.J. Zhang, K.S. Song, Y. Nakamura, T. Ueno, T. Funatsu, I. Ohdomari, H. Kawarada, Langmuir 22 (2006) 3728–3734. [33] S. Szunerits, C. Jama, Y. Coffinier, B. Marcus, D. Delabouglise, R. Boukherroub, Electrochem Commun. 8 (2006) 1185–1190.

[34] P. Actis, A. Denoyelle, R. Boukherroub, S. Szunerits, Electrochem. Commun. 10 (2008) 402–406. [35] A. Denisenko, A. Romanyuk, L.A. Kibler, E. Khon, J. Electroanal. Chem. 657 (1-2) (2011) 164–171. [36] N. Simon, C. Decorse-Pascanut, A.M. Gonçalves, D. Ballutaud, G. Charrier, A. Etcheberry, Diamond Relat. Mater. 8 (2009) 890–894. [37] S. Torrego, A. Miotello, L. Minati, I. Bernagozzi, M. Ferrari, M. Dipalo, E. Kohn, G. Speranza, Diamond Relat. Mater. 20 (2011) 990–994. [38] G. Charrier, S. Lévy, J. Vigneron, A. Etcheberry, N. Simon, Diamond Relat. Mater. 20 (2011) 944–950. [39] D. Guyomard, M. Herlem, C. Mathieu, J.L. Sculfort, J. Electroanal. Chem. 216 (1987) 101. [40] J. Turkevich, P.C. Stevenson, J. Hillier, Discuss. Faraday Soc. 11 (1951) 55–75. [41] G. Frens, Nat. Phys. Sci. 241 (1973) 20–22. [42] X.H. Ji, X.N. Song, J. Li, Y.B. Bai, W.S. Yang, X.G. Peng, J. Am. Chem. Soc. 129 (2007) 13939–13948. [43] S. Diegoli, P.M. Mendes, E.R. Baguley, S.J. Leigh, P. Iqbal, Y.R.G. Diaz, S. Begum, K. Critchley, G.D. Hammonds, S.D. Evans, D. Attwood, I.P. Jones, J.A. Preece, J. Exp. Nanosci. 1 (2006) 333–353. [44] W. Haiss, N.T.K. Thanh, J. Aveyard, D.G. Fernig, Anal. Chem. 79 (2007) 4215–4221. [45] H. Girard, N. Simon, D. Ballutaud, M. Hermemand, A. Etcheberry, Diamond Relat. Mater. 16 (2007) 888–891. [46] J.F. Moulder, W.F. Stickle, P.E. Obol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corporation, Eden Prairie, MN, USA, 1995. [47] H. Cachet, C. Debiemme-Chouvy, C. Deslouis, A. Lagrini, V. Vivier, Surf. Interface Anal. 38 (2006) 719–722. [48] K. Kishi, Y. Ehara, Surf. Sci. 176 (1986) 567–577. [49] G. Chamoulaud, D. Bélanger, J. Phys. Chem. 111 (2007) 7501–7507. [50] S. Griveau, D. Mercier, C. Vautrin-Ul, A. Chaussé, Electrochem. Commun. 9 (2007) 2768–2773. [51] T. Sainsbury, T. Ikuno, D. Okawa, D. Pacilé, J.M.J. Fréchet, A. Zettl, J. Phys. Chem. 111 (2007) 12992–12999. [52] D. Aureau, Y. Varin, K. Roodenko, O. Seitz, O. Pluchery, Y.J. Chabal, J. Phys. Chem. 114 (2010) 14180–14186. [53] S. Szunerits, M. Manesse, P. Actis, B. Marcus, G. Denuault, C. Jama, R. Boukherroub, Electrochem. Solid-State Lett. 10 (2007) G43–G46. [54] S. Szunerits, C. Jama, et al., Electrochem. Commun. 8 (2006) 1185–1190. [55] M.S. Hybertsen, L. Venkataraman, J.E. Klare, A.C. Whalley, M.L. Steigerwald, C. Nuckolls, J. Phys. Condens. Matter 20 (2008) 374115. [56] R. Sardar, A.M. Funston, P. Mulvaney, R.W. Murray, Langmuir 25 (2009) 13840–13851. [57] M. Wang, N. Simon, G. Charrier, M. Bouttemy, A. Etcheberry, M. Li, R. Boukherroub, S. Szunerits, Electrochem. Commun. 12 (2010) 351–354.