Characterization of electrodeposited nickel coatings from sulphamate electrolyte without additive

Characterization of electrodeposited nickel coatings from sulphamate electrolyte without additive

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3 available at www.sciencedirect.com www.elsevier.com/locate/matchar Characteriz...

1008KB Sizes 0 Downloads 5 Views

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3

available at www.sciencedirect.com

www.elsevier.com/locate/matchar

Characterization of electrodeposited nickel coatings from sulphamate electrolyte without additive A. Godon a , J. Creus a , X. Feaugas a , E. Conforto b , L. Pichon c , C. Armand d , C. Savall a,⁎ a

Laboratoire d'Etudes des Matériaux en Milieux Agressifs, EA3167, Université de La Rochelle, Av. Michel Crépeau, F-17042 La Rochelle, France b Fédération de Recherche en Environnement pour le Développement Durable (FR-EDD), FR CNRS 3097, Centre Commun Analyses, Université de La Rochelle, 5 Allée de l'Océan, F-17042 La Rochelle Cedex 9, France c Institut Pprime, UPR 3346 CNRS, Université de Poitiers, SP2MI, Boulevard Marie et Pierre Curie, BP 30179, 86962 Chasseneuil, Futuroscope Cedex, France d INSA Toulouse, Département de Physique, 135 avenue de Rangueil, 31077 Toulouse Cedex 4, France

AR TIC LE D ATA

ABSTR ACT

Article history:

In this paper, the influence of deposition current density on microstructure and purity of

Received 28 June 2010

nickel coatings was studied. Complementary characterization methods (SEM, TEM, XRD,

Received in revised form

EBSD, GDOES and SIMS) were used to investigate different scales of the microstructure and

25 October 2010

to understand the metallurgical states of the coatings. As deposition current density

Accepted 18 November 2010

decreases, grain refinement and texture modifications are observed which are linked with the grain boundary character (disorientation angle and Coincidence Site Lattice). Moreover,

Keywords: Electrodeposited nickel Grain refinement

in sulphamate bath without additive, the contamination by light elements and metallic impurities strongly depends on deposition parameters and must be taken into account to discuss the microstructure changes. © 2010 Elsevier Inc. All rights reserved.

Grain boundaries EBSD Chemical composition

1.

Introduction

Nanocrystalline materials have been the subject of intensive research because of their unique properties [1–3]. For example, concerning the corrosion resistance of pure metals, several works report that the susceptibility to localized corrosion is lower in nanocrystalline materials [4–6], but the mechanisms responsible for this superior corrosion resistance are not clearly established [5,7]. As corrosion resistance can be affected by several metallurgical parameters (defects, grain size, grain boundary, purity, crystallographic texture, roughness, etc.), a careful control of microstructure is necessary.

Nanocrystalline nickel with a grain size below 100 nm was obtained by electrodeposition but deposition parameters largely vary from one study to another. For example, in additive-free Watts bath [8] ultra-fine-grained nickel electrodeposits (grain size down to 100 nm) were obtained by pulse plating at very high pulse-current. By using organic additives (especially saccharin in the case of nickel), several studies show that it was possible to produce nanocrystalline nickel coatings in different baths with grain sizes in the range of 6– 100 nm [6,8,9]. It was shown that the use of organic additives leads to an increase of the contamination of coatings [10,11], which can affect both mechanical properties and corrosion

⁎ Corresponding author. Tel.: +33 5 46 45 72 93; fax: + 33 5 46 45 72 72. E-mail address: [email protected] (C. Savall). 1044-5803/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.matchar.2010.11.011

M A TE RI A L S CH A RACT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4 –1 7 3

resistance. Few studies have tried to explain the influence of deposition parameters by a careful analysis of the microstructure [8,12]. In most cases, only one parameter is studied, mainly grain size which is evaluated by analyzing the broadening of the diffraction peaks [3,9,13] or by scanning electron microscopy [14,15]. However, it was shown that for the same electrodeposited nickel sample, the size of structural elements can largely vary depending on the observation tool, and thus the microstructure needs to be evaluated at different scales [11]. Among the different baths, sulphamate based bath is of particular interest as it leads to ductile deposits with low internal stress [17,18], even without sulphur (S) containing additive [16]. In this paper, an additive-free sulphamate bath is used in order to limit the incorporation of impurities and especially S because of its dramatic effect on corrosion resistance. The influence of current density on the microstructure and on contamination of nickel coatings is studied by using different characterization methods. The correlation between structural observations at different scales and chemical analysis allows understanding the metallurgical states of the coatings.

2.

Material and Methods

Nickel coatings were prepared by direct-current galvanostatic deposition in a three-electrode cell by using a VSP potentiostat from Biologic. A conventional sulphamate bath (V = 400 mL) without additive was used, and composed of 300 g/L Ni(NH2 SO3)2.4H2O, 15 g/L NiCl2.6H2O, and 30 g/L H3BO3. Solutions were prepared by dissolving pure salts in ultrapure water (18.2 MΩ cm) and pH was then adjusted to 4.2 by adding nickel carbonate. Special attention was devoted to avoid contamination of the bath. A thermostated glass reactor was used to fix the temperature at 50 °C and the solution was mechanically stirred during the deposition. The anode was of pure nickel (99.99%) and embedded in a polypropylene anode bag. Nickel substrates (S = 2 cm2) were polished with silicon carbide (particle size 5 μm), sonicated for 2 min, rinsed with ultrapure water and dried before electrodeposition. Deposition current density was varied between 1 and 50 mA/cm2. In the following, the nomenclature of samples (Table 1) refers to this deposition parameter (for example CD 1 refers to a deposition current density of 1 mA/cm2). Cathodic efficiency was estimated by weighting the samples before and after deposition. Deposition time was adjusted to obtain thicknesses of 50 μm.

Table 1 – Mean sizes deduced from SEM, EBSD and TEM for coatings elaborated at different current densities. Name

j mA/cm2

Φ (SEM) μm

d (EBSD) μm

d (TEM) μm

CD1 CD5 CD10 CD20 CD50

1 5 10 20 50

0.37 0.74 1.4 3.9 4.3

0.25 0.35 – – 1.02

0.120 0.180 – – –

165

The surface morphology was observed by scanning electron microscopy (SEM) with a FEI Quanta 200 ESEM-FEG operating at 20 kV as acceleration voltage. Electron backscatter diffraction (EBSD) was used to obtain grain size and to characterize microtexture and grain boundaries. For top-view EBSD analyses, samples of 75 μm thickness were electrodeposited and then electropolished in a H2SO4/CH3OH mixture [19] in order to remove 25 μm. After electropolishing, samples were very flat, with a roughness below 2 nm (estimated by Atomic Force Microscopy experiments). For cross-section EBSD analyses, samples were cut with a wire saw and cross-sections were mechanically polished. A final polishing was performed with OPS preparation from Struers. EBSD maps were acquired at half of the coating thickness using an acceleration voltage of 25 kV on SEM and the TSL OIM Data collection 5 Software, with a step size of 30 nm or 70 nm, depending on the grain size. A clean-up was performed on maps in order to remove points which were not indexed or to index according to the first neighbours those which were originally incorrectly indexed. Grain size and orientation pictures were then calculated using TSL OIM Analysis 5 software. Complementary transmission electronic microscopy (TEM) observations were carried out with a JEOL JEM 2011 electron microscope operating at 200 kV. Foils for TEM were thinned in double twin-jet electro-polisher using an electrolyte of 25% nitric acid and 75% methanol at a temperature of 30 °C and a current of 150 mA. To understand the microstructure observed at high current density, TEM observations were also performed on the cross-sections of sample CD50. For this specimen, stereographic analyses (stereographic projection) were established for each observed grain in order to evaluate the orientation of each grain. Special care was taken in the marking of TEM specimens. So, the direction of the normal of the electrodeposited surface was identified on the stereographic map of each studied grain. X-ray diffraction analyses in θ–2θ mode were performed on a Brucker apparatus (AXS D8-Advanced) with the Cu-Kα radiation (λ = 0.15405 nm). Spectra were acquired between 40° and 100°, with a step width of 0.02° and the Kα2 peak and background were removed. Composition analyses were obtained by Glow Discharge Optical Emission Spectrometry (GD Profiler from Horiba Jobin Yvon). Secondary Ion Mass Spectrometry (IMS 4FE6 from CAMECA) was also used with two ionic sources Cs+ (at 14.5 keV) and O+2 (at 5.5 keV) to obtain the best sensitivity. Concentration profiles were acquired after a pulverization of 5 to 10 μm in order to avoid surface contamination effects. All atomic elements were analyzed except nitrogen. For both methods, the detection limit for this element was too high. Calibration with bulk nickel samples of known composition was performed for quantitative analysis. Several profiles were obtained for each sample, leading to reliable results. However, due to the small volumes which are analyzed by these techniques, concentration values cannot be given with a high accuracy.

3.

Results

The aim of this work is to use complementary analyses to obtain an overview of the metallurgical state of electrodeposited

166

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3

coatings. Section 3.1 describes surface morphologies in relation with SEM observations. Section 3.2 outlines the interest to use X-ray diffraction analyses to study the macroscopic texture and to extract dimensional data. The following Sections 3.3 and 3.4 deal with the opportunity to obtain spatial information using EBSD maps and TEM analyses. Finally, chemical composition was analyzed in connection with structural results.

3.1.

Scanning Electron Microscopy

SEM views presented in Fig. 1 show the surface morphology of coatings prepared at different current densities. At high current densities, large crystallites with a truncated pyramidal shape are observed leading to bright deposits in good agreement with previous results in sulphamate bath [12,15]. A strong hydrogen evolution leads to the formation of bubbles and edge effects at current densities above 50 mA/cm2. As the current density decreases, this pyramidal morphology is replaced by a nodular morphology. The mean size of the nodules deduced from SEM was estimated by statistical analyses of images obtained at different magnifications (Table 1). The values suggest a refinement at low current density. However, the morphological features observed by

SEM cannot be directly assigned to grains and other characterization tools will be used below to clarify this point.

3.2.

X-ray Diffraction Analysis

The diffraction patterns for different deposition current densities are plotted on Fig. 2. At high current density (above 30 mA/cm2), a strong crystallographic texture along the <100> direction is observed, which is replaced by a <110> preferred orientation at current densities below 20 mA/cm2. At 1 mA/cm2, no preferred orientation is observed but the (220) line is slightly high and the (200) one is slightly low respectively to a non texture nickel sample (JCPDS data no. 00-004-0850). Complementary texture analysis by using inverse pole figures obtained by EBSD will be presented in Section 3.3, confirming the above results. For coating CD1, a broadening of the diffraction peaks can be noticed, suggesting a grain refinement effect. Assuming a Cauchy-shaped profile, the full width at half maximum (FWHM) was evaluated for each diffraction peak, after correction by the experimental broadening estimated by using the LaB6 standard sample. The Scherrer equation obviously led to a strong underestimation of the grain sizes of these coatings. So, an approach based on

Fig. 1 – SEM top views showing the surface morphology of the coatings deposited at different current densities. (a: CD1 (1 mA/cm2), b: CD5 (5 mA/cm2), c: CD10 (10 mA/cm2), and d: CD50 (50 mA/cm2)).

M A TE RI A L S CH A RACT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4 –1 7 3

Fig. 2 – θ–2θ scans of coatings elaborated at different current densities.

Williamson–Hall diagrams was used in order to estimate crystallite sizes and internal stresses. The approach developed by Reimann [20] and used by Thiele [11] in electrodeposited nickel was followed, which takes into account the elastic anisotropy of nickel. The Williamson–Hall plots obtained for coatings deposited at 1 mA/cm2 (CD1) led to a mean internal stress (<σ2>1/2) of 300 MPa. This value is in the range of those previously reported in electrodeposited nickel [11] which showed an increase of mean internal stress as the grain size decreases. The mean size of coherent scattering regions for sample CD1 deduced from this analysis is around 130 nm. For coatings deposited at higher current densities, the broadening of the diffraction peaks is smaller. Moreover, for these coatings, the presence of a crystallographic texture does not allow this approach.

3.3.

Electron Backscatter Diffraction

Top-view orientation maps for coatings elaborated at different current densities are presented on Fig. 3. Inverse pole figures were calculated from these orientation maps, showing the orientation densities for the different crystallographic directions parallel to the sample normal direction. The preferred orientation along the <100> direction suggested by θ–2θ XRD scans for coatings prepared at 50 mA/cm2 (CD50) is confirmed. Comparison with SEM views shows that the large truncated pyramidal structures are mainly oriented with their <100> axis perpendicular to the substrate surface. Between these pyramidal grains, much smaller grains are found, with different crystallographic orientations. Even if a preferred orientation along the <110> direction is found for the coating prepared at 5 mA/cm2 (CD5), the texture is less marked (as the proportion of pixels which <110> crystal direction is disoriented versus the sample normal direction is higher). For the coating deposited at the lowest current density (CD1), the crystallographic texture along the <110> direction is very weak, in accordance with θ–2θ XRD scans. Grain boundary position is superimposed as grey lines to the orientation maps of Fig. 3. Neighbouring pixels in the map with disorientation smaller than 5° are associated with the same grain. According to this

167

disorientation angle, the grain size distribution can be measured and a mean grain size (dEBSD) can be evaluated. For each sample, the analyzed area was large enough to take into account more than 5000 grains. The results are given in Table 1, and in accordance with SEM observations, the grain size decreases and the grain distribution becomes narrower when the deposition current density is reduced. EBSD orientation maps obtained on cross-sections of different deposits are given on Fig. 4. The growth direction which is perpendicular to the surface of the substrate is also shown on this figure. For the CD50 sample, fibers (whose axis is perpendicular to the substrate surface) characterized by a dominant colour are observed. These fibers are formed by grains slightly disoriented with regard to the neighbouring grains, but with the (100) direction mainly parallel to the growth direction. Between these fibers, some less oriented regions are found. The thickness of these fibers (around 5 μm) is quite similar to the size of large crystallites with a truncated pyramidal shape, which are observed on the surface (4.3 μm, Table 1). As the deposition current density decreases, these fibers are no longer observed and the mean size of the grains decreases. It can be noticed that the grains do not show any elongation along the growth direction whatever the deposition current density. Two parameters are mainly used to describe the nature of grain boundaries: the disorientation angle and the Σ factor, which denotes the fraction of atoms in the grain boundary plane which are coincident to both lattices. These parameters were evaluated by using EBSD [21] and are given in Table 2 and Fig. 5. An increase of the fraction of high angle grain boundaries (HAGB) is observed as the grain size decreases and as the marked texture along the <100> direction is replaced by a weak texture along the <110> direction (Table 2). The amount of coincidence site lattice (CSL) is also strongly modified, showing a decrease of the abundance of Σ1 boundaries and an increase of the number of Σ3 and Σ9 boundaries when the grain size decreases (Fig. 5).

3.4.

Transmission Electronic Microscopy

Grain size was evaluated using TEM observation on a population around 150 grains and the mean values are given for CD1 and CD5 in Table 1. These values are lower than the ones obtained by EBSD, but for the CD1 sample, the value is in agreement with XRD analysis (130 nm). As a strong heterogeneity of grain sizes was observed for CD50, the mean value is not relevant for this sample. TEM observations were also performed on cross sections for this sample to evaluate the crystallographic orientation of different grains. These analyses are time consuming, thus only a semi-statistical study on 56 grains at different locations inside the sample was performed. However, 56 grains seemed to be sufficient to reflect the heterogeneity of the sample, as the results were not significantly modified when this number was increased. Different populations of grains were identified, characterized by three angles ψ(100), ψ(111) and ψ(110) (Fig. 6). ψ(hkl) relates the angle between (hkl) plane and the normal to the coating surface. The first one (V1) corresponds to the largest grains (>700 nm) and exhibits an angle ψ(100) near 0. This means that this crystallographic population mainly contributes to the

168

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3

Fig. 3 – Left: Top view orientation maps obtained by EBSD for coatings elaborated at different current densities: CD1 (a), CD5 (b), and CD50 (c). Right: Inverse pole figures of the normal direction for the three coatings.

macroscopic texture observed by XRD. In a “random” zone (cf. EBSD analyses), three other kinds of crystallographic populations were identified (Fig. 6), which do not correspond to macroscopic texture obtained by XRD. The size of these grains is generally lower (130 to 250 nm) than the grain with V1 variant. The correlations of these observations with SEM and EBSD results show that two kinds of regions can be distinguished in the CD50 coating: the first one corresponds to large

grains with a <100> preferred orientation and the second one is associated with “random” regions, with a much lower grain size and weaker texture.

3.5.

Composition Analysis

Table 3 lists the different elements detected in the coatings and their contents in weight ppm obtained by SIMS and

M A TE RI A L S CH A RACT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4 –1 7 3

169

Fig. 4 – Cross-section orientation maps obtained by EBSD for coatings elaborated at different current densities: CD1 (a), CD5 (b), and CD50 (c). The substrate surface normal is given by an arrow.

GDOES. For the coating CD50, impurity amounts are very low, leading to a purity around 99.99%. However, for the coatings prepared at lower current density the contamination drastically increases especially for light elements (H, O, C, etc.) and for Cl and Cu. For these coatings, some impurity contents are given with a large inaccuracy, and the purity of the coating could not be evaluated. In these cases and especially for chloride for which the concentration in ppm was not given, the quantification was not reliable as the reference samples contained much lower amounts of these elements. Concentration profiles and cartographies were obtained for each atomic element, showing that the impurities were homoge-

neously distributed laterally and through the thickness of the coatings.

4.

Discussion

Electrodeposited layers often exhibit a fiber texture, i.e. preferred crystallographic orientation of their crystallites along the growth direction, which is the case for deposits CD5 and CD50. Our results are in good agreement with published results for sulphamate bath which report a strong crystallographic texture along the <100> axis associated

170

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3

Table 2 – Classification of the grain boundaries character for coatings deposited at different current densities. Lowangle grain boundaries (LAGB) are characterized by a disorientation angle below 15° and high-angle grain boundaries are characterized by a disorientation angle above 15°. CD1 LAGB/HAGB (%) LCSL/HCSL (%)

CD5

CD50

7

93

9

91

42

58

42

58

57

43

57

43

CSL denotes coincidence site lattice (CSL) with low sigma (LCSL, Σ < 29) or high sigma (HCSL, Σ > 29).

with large grains for deposition currents between 30 and 200 mA/cm2 [12,13]. Based on cross-section observations by optical microscopy or SEM after chemical etching, the <100> texture has been associated with the formation of long columnar grains, some of them extending across the whole thickness of the coatings (10 to 50 μm) [15,18]. Cross-section observations of the CD50 coating by optical microscopy after acidic etching show kinds of columns, parallel to the growth direction, with a width of few micrometers (Fig. 7a). EBSD and TEM observations on cross sections allow to distinguish unambiguously the grains and to evaluate their orientations. The results obtained by EBSD clearly show that these “columns” are formed by grains weakly disoriented with regard to their neighbouring, with the <100> direction perpendicular to the substrate surface. This microstructure, associated with a quite high amount of low angle grain boundaries and particularly of Σ1 boundaries explains the large disagreement between the structure size elements deduced by EBSD and SEM observations in these coating (Table 1). Results obtained by TEM confirm that the largest grains (and the more numerous) are oriented with one direction <100> parallel to the growth direction. However, a significant amount of grains, much smaller (<250 nm), is differently oriented. As the current densities decrease, grain refinement is observed associated with the evolution of the <100> texture towards a less marked <110> texture, in good agreement with published results in similar bath [18,23]. In the literature, such

Fig. 5 – Amounts of LCSL plotted versus Σ value for coatings deposited at different current densities.

texture changes are associated with a grain size decrease, obtained by using pulse plating [2,22] or organic additives [23,24]. For coatings deposited at low current densities, very thin fibers parallel to the growth direction are observed on cross-sections after chemical etching (Fig. 7b). However, EBSD observations clearly show that the grains are not elongated along the growth direction. In agreement with the results obtained for the coating CD50, EBSD is a powerful tool to observe the microstructure of electrodeposited coatings and to avoid artefacts linked with chemical etching necessary to display grains with more conventional observation techniques. EBSD analyses presented here show that these evolutions of texture and grain size are associated with an increase of the amount of high angle grain boundaries. Particularly, a decrease of the fraction of Σ1 boundaries and an increase of the fraction of Σ3 boundaries are observed as the grain size decreases. Similar trends were reported in copper proceeded by equal Channel Angular Extrusion [25]. Several studies suggest that the presence of low-Σ coincidence site lattice (CSL) boundaries could be associated with a better corrosion resistance [7,26] and with better mechanical properties [27]. However, this can be moderated by the fact that the presence of low-Σ coincidence site lattice boundaries seems to be correlated with high impurity contents. At low current densities, equiaxed nickel with very low grain sizes can be deposited in direct current mode without additive. For these deposits, good correlations between grain sizes deduced from EBSD and nodule sizes deduced from SEM are found. At low deposition current densities, the random orientation of grains is linked with high disorientations between grains, which appear as distinct entities in top-view SEM observations of the coatings. A mean grain size value around 250 nm is found by EBSD (with a disorientation of 5°) for the coating prepared at the lowest current density. A much smaller value is deduced from analyzing the broadening of the diffraction pattern in accordance with the grain size measured by TEM. Thus, the choice of the disorientation angle value used to define grain size by EBSD needs to be validated by a correlation with XRD or TEM analyses. Our results show that the grain refinement and the changes of crystallographic orientation are linked with an increase of the amounts of several impurities. As the nickel anode was of high purity, Co and Cu contamination probably originates from the chemicals of the bath. Copper, which is nobler than nickel, is preferentially deposed at low current density (and thus at low overpotential). Because of their position in the Mendeleïv table, these metallic impurities can easily replace nickel in the coating and should act as substitution impurities. The changes of the microstructure are probably linked with the incorporation of light atomic elements. Inhibition phenomena [28,29], are known to strongly influence electrocrystallisation processes. In the case of nickel electrodeposition from Watts bath, Amblard et al. [30] show that interfacial inhibitors (Hads, Ni(OH)2, etc.) led to several growth modes and textures, depending on deposition parameters, but different species including C, O, N, H or Cl atoms were proposed [29,31]. A significant drop of the deposition efficiency was observed as the deposition current density decreases from 98% (for coatings deposited at 50 mA/cm2) to around 84% (for

M A TE RI A L S CH A RACT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4 –1 7 3

171

Fig. 6 – TEM observations (CD50) and stereographic projections showing the orientation of different grains in a “random oriented” region. The table gives the grain population in terms of angle between the coating normal surface and the (hkl) plane.

coatings deposited at 1 mA/cm2). Voltammograms obtained in the plating bath with the same stirring conditions allowed us to estimate the dioxygen reduction current density around

Table 3 – Impurity content in weight ppm for coatings deposited at different current densities. For the values in italics, the quantification was not possible as the reference samples contained a much lower amount of these impurities.

CD50 CD5 CD1

H

C

O

S

Cl

Fe

Co

Cu

Mo

1 ≈ 70 ≈ 130

5 ≈ 100 ≈ 400

25 ≈ 400 ≈ 1000

<1 6 25

<1 ×230 ×1000

4 15 7

32 100 155

25 150 ≈ 600

<40 <40 <40

0.1 mA/cm2 and thus the contribution of this reaction could explain the decrease of deposition efficiency. At low current densities, and thus low deposition rates, this reaction could hinder the growth of crystallites, contributing to the refinement effect. More generally, the adsorption of different foreign species (including O, H, C, and Cl) at the cathode surface probably prevents grain growth by avoiding surface diffusion of adatoms and significant amounts of these species are incorporated into the coatings. EBSD analyses show that, in coatings deposited at low current density, grain boundaries are more defective (higher disorientation angle and Σ factor) with probably an increased concentration of vacancies. Thus, the results are consistent with a decrease of grain size when current density decreases, associated with the incorporation of impurities at grain boundaries.

172

MA TE RI A L S CH A R A CT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4– 1 7 3

and higher contamination. Both light elements and substitution impurities are incorporated when grain size decreases, which can affect mechanical properties and corrosion resistance. So, chemical contamination of electrodeposited coatings must be carefully evaluated before discussing the influence of their microstructure on properties.

Acknowledgement Thanks are due to the Agence Nationale de la Recherche (GIP ANR Program no. ANR-06JCJC-0023-01) for the financial support.

REFERENCES

Fig. 7 – a) Cross-section view of the coating CD50 obtained by optical microscopy after chemical etching, and b) cross-section view of the coating CD5 obtained by SEM after chemical etching.

5.

Conclusion

Although extensive experimental works have been published concerning characterization of nickel electrodeposited coatings, a study combining composition analyses and multi-scale microstructural characterization is missing. In sulphamate bath without additive, microstructure modifications are linked with the incorporation of impurities and particularly light atomic elements whose content largely depends on electrodeposition conditions. Deposits obtained at current densities above 20 mA/cm2 show a strong <100> texture along the growth direction but are characterized by different structural heterogeneities which can be evidenced by using complementary observation tools. TEM and EBSD observations offer the opportunity to distinguish the different microstructural scales and to better understand the microstructure of coatings. As the current density decreases, grain refinement and texture modifications are observed which are associated with more defective grain boundaries

[1] Qin L, Xu J, Lian J, Jiang Z, Jiang Q. Surf Coat Technol 2008;203: 142–7. [2] Qu NS, Zhu D, Chan KC, Lei WN. Surf Coat Technol 2003;168: 123–8. [3] Rashidi AM, Amadeh A. Surf Coat Technol 2008;202:3772–6. [4] Kim SH, Erb U, Aust KT, Gonzalez F, Palumbo G. Plat Surf Finish 2004;91:68–70. [5] Miyamoto H, Harada K, Mimaki T, Vinogradov A, Hashimoto S. Corros Sci 2008;50:1215–20. [6] Mishra R, Balasubramaniam R. Corros Sci 2004;46:3019–29. [7] Roy I, Yang HW, Dinh L, Lund I, Earthman JC, Mohamed FA. Scr Mater 2008;59:305–8. [8] El Sherik AM, Erb U, Page J. Surf Coat Technol 1996;88: 70–8. [9] Ebrahimi F, Ahmed Z. J Appl Electrochem 2003;33:733–9. [10] Wang YM, Cheng S, Wei QW, Ma E, Nieh TG, Hamza A. Scr Mater 2004;51:1023–8. [11] Thiele E, Klemm R, Hollang L, Holste C, Schell N, Natter H, et al. Mater Sci Eng A 2005;390:42–51. [12] Rasmussen AA, Moller P, Somers MAJ. Surf Coat Technol 2006;200:6037–46. [13] Zhao H, Liu L, Zhu J, Tang Y, Hu W. Mater Lett 2007;61: 1605–8. [14] Xuetao Y, Yu W, Dongbai S, Hongying Y. Surf Coat Technol 2008;202:1895–903. [15] Banovic SW, Barmak K, Marder AR. J Mater Sci 1998;33: 639–45. [16] Kelly JJ, Goods SH, Talin AA. Electrochem Soc Proc 2004;17: 432–47. [17] Baudrand D. Met Finish 1996;94:15–8. [18] Marquis EA, Talin AA, Kelly JJ, Goods SH, Michael JR. J Appl Electrochem 2006;36:669–76. [19] Sahal M, Creus J, Sabot R, Feaugas X. Acta Mater 2006;54: 2157–67. [20] Reimann K, Würschum R. J Appl Phys 1997;81:7186–92. [21] pdf database EBSD, OIMDC Manual, Copyright 1997–2006, EDAX-TSL. [22] Fritz T, Cho HS, Hemker KJ, Mokwa W, Schnakenberg U. Microsyst Technol 2003;9:87–91. [23] Ebrahimi F, Bourne DG, Kelly MS, Matthews TE. Nanostruct Mater 1999;11:343–50. [24] Lin CS, Hsu PC, Chang L, Chen CH. J Appl Electrochem 2001;31: 925–33. [25] Dalla Torre FH, Gazder A, Gu CF, Davies CHJ, Pereloma EV. Metall Mater Trans A 2003;38:1080–95. [26] Palumbo G, Aust KT, Lehockey EM, Erb U, Lin P. Scr Mater 1998;38:1685–90.

M A TE RI A L S CH A RACT ER IZ A TI O N 62 ( 20 1 1 ) 1 6 4 –1 7 3

[27] Godon A, Creus J, Cohendoz S, Conforto E, Feaugas X, Girault P, Savall C. Scr Mater 2010;62:403–6. [28] Fischer H. Electrod Surf Treat 1972/73;1:319–34. [29] Winand R. Electrochim Acta 1994;39:1091–105.

[30] Amblard J, Epelboin I, Froment M, Maurin G. J Appl Electrochem 1979;9:233–42. [31] Natter H, Schmelzer M, Hempelmann R. Mater Res Soc 1998;13:1186–97.

173