Jialei Zhangab,
Changdong Gu*ab,
Yueyu Tongab,
Junming Goua,
Xiuli Wangab and
Jiangping Tuab
aSchool of Materials Science and Engineering, State Key Laboratory of Silicon Materials, Zhejiang University, Hangzhou 310027, China. E-mail: cdgu@zju.edu.cn; changdong_gu@hotmail.com; Fax: +86 571 87952573; Tel: +86 571 87952573
bKey Laboratory of Advanced Materials and Applications for Batteries of Zhejiang Province, Hangzhou 310027, China
First published on 10th August 2015
Cr and Cr–P coatings were electrodeposited on Fe substrates from non-aqueous deep eutectic solvent-based electrolytes containing Cr(III). The optimized deposition parameters for the coating process were explored. A two-step process of Cr(III) reduction occurred, i.e. Cr(III) → Cr(II) → Cr(0), and the controlling step was promoted by adding NH4H2PO2. It was found that an electro-brush plated Ni underlayer was essential to obtain a smooth and compact Cr or Cr–P coating on the Fe substrate. The structure and composition of the as-deposited coatings were thoroughly analyzed. The corrosion behavior of the Cr and Cr–P coatings is quite different in 3.5 wt% NaCl and 0.1 M H2SO4 solutions. The diplex effects of the layered structures and ion-selective components in the as-prepared Cr-based coatings are suggested to be responsible for the different corrosion mechanism in different corrosion media.
As a promising alternative to Cr(VI) in the electroplating process, Cr(III) has been received significant attention in recent decades. The less toxic Cr(III) occurs in many foods, and some dietary intake is essential to human health.9 Developments and improvements in environmentally-friendly Cr(III) electroplating technology have become a hot research area in the 21st century. Nowadays, many problems have arisen in the burgeoning Cr(III) electroplating process. For instance, the electrochemical reduction mechanism of trivalent Cr remains unclear.9,10 It is difficult to improve the thickness of Cr from a Cr(III) plating bath.11 The intermediate Cr3+ is unstable and easily oxidized to Cr6+ near the anode during electroplating.12 It is hard to find a suitable complex agent or electrolyte solvent for Cr(III) electrodeposition, because it is almost impossible to deposit Cr coatings from a simple aqueous Cr(III) solution due to the very stable [Cr(H2O)6]3+ complex.13–15 Moreover, considerable hydrogen evolution cannot be avoided during the reduction of Cr3+ to Cr0.14,15
Nevertheless, previous efforts in Cr(III) aqueous plating research reveal that high-quality Cr coatings can be electrodeposited from electrolytes containing complex ions, which are constituted by trivalent Cr and organic acid complexants possessing fewer than six carbon atoms.10,16,17 Inevitably, the final Cr coatings electrodeposited from these baths usually have a high carbon content due to the reduction of organic acids.18 Meanwhile, great attention has also been focused on the electrodeposition of Cr(III) from ionic liquids, which have the nature of multicomponent ions and complex organic species, to facilitate coordination chemistry or proton transfer.19,20 In other words, ionic liquids themselves could provide organic complexing agents for Cr(III) electrodeposition. Most investigations for non-aqueous Cr(III) electrodeposition have mainly focused on 1-butyl-3-methylimidazolium ([BMIM])-based ionic liquids.2,21–25 In addition, a few researchers have reported that chromium could be successfully electrodeposited from a relatively cheap choline chloride-based deep eutectic solvent (DES).8,26–28 On the other hand, the co-deposition of chromium and phosphorus has seldom been explored.3,13,18,29,30 The addition of a phosphorus metalloid to chromium may lead to better corrosion resistance.18,30 The main challenges for electrodepositing a Cr–P alloy are accompanied by the problems of Cr(III) electrodeposition as mentioned above.
In this work, we propose a Cr(III) electroplating process using a DES-based electrolyte to produce Cr coatings on Fe substrates. A Cr–P alloy coating was also co-electrodeposited by adding ammonium hypophosphite (NH4H2PO2) as a phosphorus source. The as-prepared Cr and Cr–P coatings consist mainly of Cr-based compounds containing Cr, Cr2O3 and Cr(OH)3. The inevitable presence of oxides and hydroxides, which result from the additional crystalline hydrate and surface oxidation, greatly influences the corrosion behavior of the as-deposited Cr and Cr–P coatings in NaCl and H2SO4 aqueous solutions.
A Ni underlayer with a thickness of ∼2 μm was prepared on the iron sheet (0.2 mm in thickness with Fe > 98.576% and C < 0.025%) from a 0.5 M NiCl2-DES electrolyte by electro-brush plating at 4 V, as in our previous work.31 The Cr and Cr–P coatings were then direct-current electrodeposited on the Fe/Ni from the above electrolytes at a constant potential of −1.2 V in a three-electrode electrochemical workstation (CH Instruments, Inc., China). Graphite was used as the anode and Ag wire was used as a quasi-reference electrode. The electrolyte temperature was maintained at 55 °C on a magnetic stirrer with a stirring rate of 800 r min−1 during the electrodeposition. After deposition for 30 min the Fe/Ni/Cr and Fe/Ni/Cr–P deposits were sequentially rinsed with methanol and deionized water and then dried under nitrogen flow.
The crystalline structure of the deposit was investigated by X-ray diffraction (XRD, XPert Pro-MPD with CuKα radiation, λ = 0.15406 nm). The surface morphology was observed using a field emission scanning electron microscope (FE-SEM, Hitachi SU-70). The surface topography and roughness were measured using an atomic force microscope (AFM, Nanosurf NaioAFM, Switzerland) in contact mode. An energy dispersive X-ray spectrometer (EDS) was used for qualitative elemental analysis of the Cr and Cr–P surfaces. X-ray photoelectron spectroscopy (XPS, Axis Ultra DLD) was further employed to analyze the chemical composition and state of the coatings using AlKα (monochromatic) radiation with E = 1486.6 eV. All the core level spectra were referenced to the C 1s peak at 284.8 eV.
Potentiodynamic polarization measurements were carried out in 3.5 wt% NaCl and 0.1 M H2SO4 aqueous solutions at room temperature using a three-electrode cell (CH Instruments, Inc., China) with a platinum plate as the counter electrode and a Ag/AgCl electrode as the reference electrode. The specimens were covered by polyimide coatings, leaving an exposed area of ∼1 cm2, and used as working electrodes. Each specimen was immersed into the solutions for about 20 min to stabilize the open-circuit potential. The corrosion potential (Ecorr) and corrosion current density (icorr) were obtained by the Tafel extrapolation method from the potentiodynamic polarization curves with a scan rate of 1 mV s−1.
Fig. 1b shows the CV curves of the DES-based electrolytes containing CrCl3 and CrCl3 + NH4H2PO2 at 55 °C. The potential scan started from 1.0 V (vs. Ag wire), proceeding in the negative potential direction up to −1.5 V, and was then reversed to the starting potential at a scan rate of 20 mV s−1. The CrCl3-electrolyte reveals two pairs of redox peaks, which correspond to the two consecutive steps of Cr(III) reduction, i.e. Cr(III) → Cr(II) → Cr(0).8,16,34 In the cathodic branch, with increasing negative potential a reduction peak, Ac, is observed at about −0.75 V, which corresponds to the reduction of Cr(III) to Cr(II). At around −1.13 V the current density rises steeply (peak Bc), which may be attributed to the formation of metallic Cr along with the decomposition of the electrolyte. Qualitatively, the onset potentials of reduction corresponding to Cr(III)/Cr(II) and Cr(II)/Cr(0) in the choline chloride–ethylene glycol DES are more positive than those in the choline chloride–chromium trichloride DES.8,26 In the reverse scan, the anodic peaks Ba and Aa can be assigned to the oxidation of Cr(0) to Cr(II) and Cr(II) to Cr(III), respectively. In addition, the presence of a crossover occurring at about −1.0 V is a sign of a nucleation and growth process.35 The addition of ammonium phosphite to the CrCl3-electrolyte has a remarkable influence on the kinetics of Cr(III) electrodeposition. The decreased current density is in agreement with the decrease in conductivity as shown in Fig. 1a. In the cathodic branch, it is observed that the onset potential of Cr(III)/Cr(II) shifts from 0.00 V towards the negative region (−0.10 V) while that of Cr(II)/Cr(0) shifts from −1.13 V towards the positive region (−0.95 V). These features indicate that NH4H2PO2 can inhibit the reduction process of Cr(III) to Cr(II), but promote the formation of Cr(0) from Cr(II). Accordingly, the anodic peak corresponding to Cr(II)/Cr(0) is positively shifted (−0.55 V to −0.46 V) while that corresponding to Cr(III)/Cr(II) is negatively shifted (0.14 V to −0.18 V), which suggests an inhibition of Cr(0) to Cr(II) and a promotion of Cr(II) to Cr(III) conversion. As a result, the range between cathodic peak and anodic peak is narrowed by adding NH4H2PO2, which means the polarization is reduced. It has been reported that most Cr3+ complexes undergo stepwise charge transfer at the electrode, in which the second step of Cr2+ reduction is slow.29 Therefore, with the addition of NH4H2PO2 the kinetics of Cr(III) electrodeposition are significantly enhanced. The absence of the crossover during the negative and reverse scans for the electrolyte with NH4H2PO2 results from the absence of nucleation and growth. This phenomenon is attributed to the formation of chromium hydroxides or chromium oxides occupying the sites of nucleation, thus blocking Cr deposition on the surface of the electrode.36
A large number of experiments in this work reveal that the Cr(III) electrodeposition in this system is very sensitive to the deposition parameters. Most of them are related to the deposition potential (controlling nucleation kinetics) and current density (controlling growth rate). Fig. 1c shows the current–time curves during Cr(III) electrodeposition on the Fe substrate using varying potentials at 55 °C with a stirring rate of 800 r min−1. Three potentials of −1.1, −1.2 and −1.4 V (vs. Ag wire) were tested for the Cr(III) deposition. No film was obtained at −1.1 V because this potential is a little more positive than −1.13 V, and thus cannot lead to the reduction of Cr(II) to Cr(0), as discussed in Fig. 1b. At a potential of −1.4 V, the film is pulverous and can be easily detached from the substrate by rinsing with water. Thus, the allowed deposition potentials for an acceptable chromium film are around −1.2 to −1.3 V. The current density was adjusted by adding NaCl or changing the temperature and stirring rate at −1.2 V. It was found that the cathodic current density range for chromium film formation is around 7 to 10 mA cm−2. High-quality chromium cannot be electrodeposited at −1.2 V if the current density is not in the correct range. The addition of NaCl, applied stirring rate of 800 r min−1 and temperature of 55 °C are appropriate for obtaining the current density range for Cr and Cr–P deposition at −1.2 V.
It is reported that thick Cr coatings cannot be readily obtained from Cr(III) electrolytes.16 The as-deposited Cr and Cr–P coatings in this work are very thin, with average thicknesses of 700 nm and 500 nm, respectively. The surface morphology of the deposit therefore must be strongly influenced by the roughness of the substrate. Fig. 3 shows the typical three-dimensional AFM images (5 μm × 5 μm) of the Fe/Ni, Fe/Cr, Fe/Ni/Cr and Fe/Ni/Cr–P coatings. The surface roughness of the electro-brush plated Ni is ∼6 nm, as shown in Fig. 3a, which is much smaller than the anode-treated Fe substrate (∼55 nm, not given here).31 The Fe/Cr coating has a roughness of ∼52 nm, which is close to the Fe substrate. Deep voids can be observed in the Fe/Cr coating in Fig. 3b and the defective surface is consistent with the SEM image in Fig. 2c. Much smoother Cr and Cr–P coatings are deposited on the electro-brush plated Ni underlayer, and the roughnesses are measured to be ∼12 nm and 8 nm, respectively. Large clusters and small particles can be observed in the Fe/Ni/Cr coating, whereas only small particles can be found in the Fe/Ni/Cr–P coating. The morphology refinement of the Fe/Ni/Cr–P coating correlates with the promotion of the nucleation rate of Cr(II) → Cr(0) and inhibition of crystal growth rate by low current density when adding NH4H2PO2, as discussed for the CV curves in Fig. 1b. The surface morphology and roughness of the as-deposited Cr or Cr–P thin film greatly depend on the roughness of the substrate.
![]() | ||
Fig. 3 Typical three-dimensional AFM images (5 μm × 5 μm) of the Fe/Ni (a), Fe/Cr (b), Fe/Ni/Cr (c) and Fe/Ni/Cr–P (d) coatings. The roughness for each coating is also given. |
Fig. 4 presents the EDS analyses of the electrodeposited Fe/Ni/Cr and Fe/Ni/Cr–P coatings. The EDS element mapping corresponding to the SEM image in Fig. 2f is also given in this figure. C, O, Cl, Fe, Ni and Cr elements are detected on the surface of the Cr and Cr–P coatings, as shown in Fig. 4a and b. The carbon signals mainly come from the conductive carbon tapes used. However, the intensity of the EDS peak corresponding to oxygen is extremely strong, and cannot only be caused by the conductive carbon tapes. This result may be ascribed to the chromium oxides and/or hydroxides in the deposits.9,10 The impurity of chlorine in the deposits comes from the chromium chloride plating bath, and is detrimental to its pitting corrosion resistance.3 About 6 wt% content of phosphorus is detected in the Fe/Ni/Cr–P coating, as shown in Fig. 4b. The introduction of phosphorus in the deposit results from the well-known induced deposition in the presence of an iron-group metal (Cr).38 Considering there are no other reduction peaks observed in the CV curve in Fig. 1b, the only reduced product should be metallic Cr. In other words, the detected phosphorus is primarily in the form of oxides or elemental solute in the Cr-based deposits. The elements Cr and P hence have a uniform distribution in the Fe/Ni/Cr–P coating, as shown in Fig. 4c and d.
![]() | ||
Fig. 4 EDS analyses of the electrodeposited Fe/Ni/Cr (a) and Fe/Ni/Cr–P (b) coatings. (c and d) EDS element mapping corresponding to the SEM image of the Fe/Ni/Cr–P coating in Fig. 2f. |
![]() | ||
Fig. 5 XRD patterns of the as-deposited Fe/Ni/Cr and Fe/Ni/Cr–P coatings. The reported PDF cards for Cr, Ni and Fe are also given in this figure. |
To gain a better understanding of the differences between the Fe/Ni/Cr and Fe/Ni/Cr–P coatings, XPS studies of the as-deposited coatings have been carried out, as shown in Fig. 6. The binding energies (BEs) in the XPS spectra are calibrated by using that of C 1s (284.8 eV). The XPS survey spectra of the Cr and Cr–P coatings demonstrate that the elements C, O, Cr, Cl and P (in the Cr–P coating) are present, which agrees well with the EDS analyses in Fig. 4. As shown in Fig. 6b, the Cr 2p3/2 peak can be resolved into three peaks at 574.1, 576.6 and 577.7 eV representing the Cr0, Cr3+ in oxides and Cr3+ in hydroxides, respectively.39–42 The inevitable presence of oxides and hydroxides from the additional crystalline hydrate and surface oxidation results in a partial presence of the metallic phase (Cr0) in the deposits. No peaks around 586.7 eV are found, demonstrating that Cr(VI) is not formed in the deposits. Chromium therefore exists in two different charge states in the as-prepared coatings. As for the Fe/Ni/Cr–P coating, a significant increase in peak area corresponding to Cr hydroxides, either in the form of Cr(OH)3 or CrOOH, is obtained, indicating a Cr(OH)3 rich Cr–P coating. The increased ratio of Cr(OH)3/Cr2O3 results in a distinct shift of the Cr 2p peaks in the oxidized state towards the high BE region, as shown in Fig. 6b. At the same time, the Cr 2p peaks corresponding to Cr0 remain at 574.1 and 583.4 eV. A similar composite Cr 2p3/2 peak at 577.47 eV was detected by Li et al. when preparing Cr–P coatings by electrodeposition from trivalent chromium electrolytes.13 Tharamani et al. found that the Cr 2p3/2 core level is higher and decreases on heat treatment at 673 K when preparing a Cr–P coating by electroless plating from trivalent chromium electrolytes, which suggests the dehydration of Cr hydroxides.30 These shifts in BEs demonstrate the essential nature of Cr hydroxides in Cr–P coatings prepared from trivalent chromium electrolytes with hypophosphite as a phosphorus source.
![]() | ||
Fig. 6 (a) XPS survey spectra of the as-deposited Fe/Ni/Cr and Fe/Ni/Cr–P coatings; (b and c) Cr 2p and P 2p XPS core level spectra of the as-deposited coatings. |
The P 2p core level spectrum for the Fe/Ni/Cr–P coating reveals three peaks at 129.1, 129.8 and 133.2 eV, as shown in Fig. 6c. The BE at 129.8 eV is assigned to elemental P and the other two BEs correspond to the positively and negatively charged states, i.e. P5+ and Pδ−.13,30 The Pδ− species at 129.1 eV may come from the CrPx formed on the surface,18,43 while the P5+ species at 133.2 eV likely results from phosphate species and/or P2O5.44
![]() | ||
Fig. 7 Potentiodynamic polarization curves for the Fe substrate, Fe/Ni, Fe/Ni/Cr and Fe/Ni/Cr–P coatings in 3.5 wt% NaCl (a) and 0.1 M H2SO4 (b) aqueous solutions at room temperature. |
Sample | Ecorr (mV vs. Ag/AgCl) NaCl | icorr (μA cm−2) NaCl | Ecorr (mV vs. Ag/AgCl) H2SO4 | icorr (mA cm−2) H2SO4 |
---|---|---|---|---|
Fe | −637 | 20.1 | −511 | 1.06 |
Fe/Ni | −458 | 14.5 | −465 | 4.45 |
Fe/Ni/Cr | −437 | 4.78 | −431 | 0.177 |
Fe/Ni/Cr–P | −482 | 13.8 | −406 | 0.121 |
The different corrosion performance for the Cr and Cr–P coatings in NaCl and H2SO4 solutions is likely a result of different corrosion mechanism. The surface morphologies of the corroded films were therefore investigated and the results are shown in Fig. 8. Accordingly, two kinds of corrosion mechanism are schematically illustrated in Fig. 9 to explain the different corrosion behavior for the as-deposited Cr and Cr–P coatings in 3.5 wt% NaCl and 0.1 M H2SO4 solutions. Sheet-like corrosion products are tightly attached to the corroded Cr and Cr–P coatings and large cracks can be observed, as shown in Fig. 8a and b. Although both the as-deposited Cr and Cr–P coatings exhibit compact and crack-free surfaces, as shown in Fig. 2d and f, defects such as internal stress and chlorine exist in the coatings. For example, the introduction of hydrate, CrCl3·6H2O, provides at least 1.8 M H2O, which usually results in the presence of hydrogen in the deposits and causes internal stress and brittleness. These defects cause initial pitting and then evolve by the following ways:
![]() | ||
Fig. 9 Schematic illustrations of the corrosion mechanism for the as-deposited Fe/Ni/Cr and Fe/Ni/Cr–P coatings in 3.5 wt% NaCl (a) and 0.1 M H2SO4 (b) solutions. |
In NaCl solution (Fig. 9a), the main aggressive ion, Cl−, can cause serious pitting corrosion, which occurs perpendicularly to the Cr or Cr–P coating along the initial pitting direction. According to the polarization curves in Fig. 7a, the Ecorr of Ni is more negative than Cr but more positive than Cr–P, which means the priority for corrosion would be Cr–P > Ni > Cr in NaCl solution. Therefore, for the Cr coating, lateral corrosion extends into the Ni underlayer and gases are subsequently formed under the Cr coating. As internal stress remains in the upper brittle Cr layer, cracks form near the initial pitting and delamination occurs along the cracks, as shown in Fig. 8a. On the other hand, since the Cr coating is rich in Cr2O3, which is anion-selective,46,47 Cl− can favorably permeate into the coating and also corrodes the Cr layer. Sheet-like particles are hence formed on the Cr surface. However, the Cr(OH)3 rich Fe/Ni/Cr–P coating is cation-selective,48 and it can prevent Cl− from directly going through, which results in more Cl− migrating into the initial pitting and aggravating it. The pitting is thereby enlarged and collapsed near the cracks, thus leading to a larger contact area between the Ni underlayer and the corrosive medium. The corrosion is therefore accelerated, as characterized by the higher current density in Fig. 7a. Meanwhile, the Cr–P coating has a more negative Ecorr than Ni and it is corroded prior to Ni, which leads to more corrosion products on the Cr–P surface, especially near the large pitting (enlarged from the initial pitting). As a result, larger and deeper pitting and more corrosion products on the surface are found in the Cr–P coating than in the Cr coating. The Fe/Ni/Cr–P coating therefore shows a worse corrosion resistance in NaCl solution. Ramezani-Varzaneh et al. also reported that the addition of P from NaH2PO2 causes the anti-corrosion performance of electrodeposited coatings in 3.5 wt% NaCl solution to deteriorate.3 In the case of the ultrathin amorphous Cr–P coating in this work, the defects introduced by P should significantly influence the corrosion performance.
As for the case of H2SO4 solution (Fig. 9b), the main aggressive ion is H+ and general uniform corrosion occurs. In this case, the initial pitting caused by defects would not be aggravated or enlarged so severely. As shown in Fig. 8c and d, several deep pittings with a size of ∼200 nm are found in the Cr coating, while numerous shallow pittings with a size of 100–300 nm are observed in the Cr–P coating. These pittings likely originate from the partial impurity of chlorine in the as-prepared coatings, which is detected by EDS and XPS as shown in Fig. 4a and b and 6a. It has been reported that a chromium coating electrodeposited from chromium chloride exhibits a worse corrosion resistance than a coating deposited from chromium sulfate, and the electrolyte nature has a significant influence on the corrosion performance of chromium coatings.3 Although Ni has a more negative Ecorr than Cr in H2SO4 solution (see Fig. 7b), compared with the bare Fe/Ni, only a few defect sites of Ni are exposed in the corrosive medium through a pitting with ∼200 nm diameter on the Cr surface. This means that the active Fe/Ni is well-protected by the thin Cr/Cr–P coating in H2SO4 solution. On the other hand, the anion-selective Cr2O3 rich Fe/Ni/Cr coating prevents H+ from further corroding the coating, and more H+ ions migrate to the initial pitting, which results in longitudinal corrosion and a small deep pitting in the Cr coating. Furthermore, the cation-selective Cr(OH)3 rich Fe/Ni/Cr–P coating acts as a buffer layer, which leads to an increase in pH value and slows the permeation of H+ in the interface range.49 Therefore, the Cr–P coating in comparison with the Cr coating has a better corrosion resistance in H2SO4 solution.
This journal is © The Royal Society of Chemistry 2015 |