Synthesis and assessment of novel anticorrosive polyurethane coatings containing an amine-functionalized nanoclay additive prepared by the cathodic electrophoretic deposition method

Abstract

Cathodic electrophoretic deposition (CEPD) was utilized to cover a mild steel cathode in an aqueous solution consisting of a newly synthesized two-component one-pack waterborne polyurethane resin. The resin mixture was composed of a hydroxyl terminated polyurethane prepolymer and a blocked isocyanate cross-linker, both contain build-in tertiary amine groups to provide the condition for formation of quaternary ammonium centers under acidic pH. Thermal treatment of the electrodeposited components led to formation of crosslinked polyurethane coatings with high gel content (80%). The coating was also modified by incorporation of a clay additive. For this purpose, Cloisite 30B, a quaternary ammonium modified clay was surface modified with (3-aminopropyl)trimethoxysilane and co-electrodeposited with base polyurethane components under optimized condition. All of the prepared coatings showed promising physico-mechanical properties including very good adhesion to the mild steel surface, excellent flexibility, high impact resistance as well as high pendulum hardness. Evaluation of the corrosion inhibition properties of the coatings with and without co-electrodeposited clay nanoparticles by the Electrochemical Impedance Spectroscopy (EIS) method revealed good corrosion protection of neat polyurethane coatings when exposed to 3.5% NaCl solution for 7 days. EIS assessment on the nanocomposite coating containing 0.5 wt% of modified clay showed corrosion resistance up to 8 GΩ after 2 h immersion in NaCl 3.5%. The time needed for the failing of the coating resistance was extended to 17 days for this sample. Also, results of the salt spray test revealed a lower sign of corrosion on the mild steel plate coated with the nanocomposite sample during 17 day exposure to NaCl 5% fog.

---

Introduction

One of the serious problems in the chemical, shipping and manufacturing industries is corrosion of metals when they are exposed to an aqueous environment. To reduce this damage, organic protective coatings are generally used as a barrier to prevent transportation of moisture and oxygen from surrounding environment to metals surface.^1–4

Due to the increased legislative restrictions on the emission of organic solvents to the atmosphere, many industries were pushed to use waterborne systems instead of traditional solvent borne coatings.⁵ To apply waterborne coatings on the metals surface, varieties of procedures can be followed,⁶ among them electrophoretic method has attracted many interests.^7–9 Electrophoretic deposition (EPD) is a simple process for the assembly of charge-carrying particles on an electrode from stable colloidal suspensions in a dc electric field. The thickness and morphologies of materials obtained via EPD can be controlled precisely through alteration of the electrochemical parameters. In addition, this method provides feasible condition for high speed deposition of coatings on different metallic substrates with complex shapes. EDP of materials can be performed by cathodic or anodic methods; however, cathodic deposition has important advantages regarding the possible materials can be deposited by this method.¹⁰

Different classes of materials from biological macromolecules like chitosan¹¹ up to organic polymeric materials such as epoxy^12,13 and acrylate^14–16 resins as well as polyurethanes^17,18 have been modified for the application as surface coatings of metallic substrates through EPD process. Due to advantages such as solvent, stain and chemical resistance, toughness with flexibility, environmental safety, good adhesion and rheology characteristics the application of waterborne polyurethane dispersions was considered as the main chemical framework of the coating materials developed in the present study.¹⁹

It is well known that the incorporation of nanoparticles can provide extensive opportunities for enhancement of coatings performance including their corrosion resistance and mechanical properties.^20–23 Therefore, inclusion of different types of nanoparticles for modification of coatings obtained via EPD process was considered in many publications.^24–27

Due to plate like shape and high aspect ratio of the clay nanoparticles they can increase the length of diffusion pathways of corrosive species when they incorporated into coating materials. In other word, they can decrease the permeability and consequently improve barrier property of coating against corrosive species like oxygen and water.^28–31 There are also some limited reports regarding incorporation of clay into the coatings during EPD process.^32–35 However, to the best of our knowledge incorporation of clay into the polyurethane based coatings via CEPD has not been reported yet.

To that end, synthesis of a novel series of two-component waterborne polyurethane system and preparation of corresponding clay containing nanocomposite via CEPD process was considered in the present study. To make clay suitable for CEPD, an organically modified clay, Cloisite 30B, was surface treated with an amino silane coupling agent. The prepared materials were fully characterized and the physico-mechanical and corrosion inhibition of neat and clay modified coatings were investigated.

Experimental

Materials

Poly(tetramethylene ether) glycol (PTMEG, M_n = 1000 g mol⁻¹) was purchased from Sigma-Aldrich and dried at 90 °C under vacuum for 24 h just before use. Isophorone diisocyanate (IPDI) from Merck was purified through vacuum distillation. Triethanolamine (TEA) and N-methyldiethanolamine (DEA) from Merck were dried at 80 °C under vacuum for 24 h. (3-Aminopropyl) trimethoxysilane (APS) from Merck was used as received. N-Methylaniline (NMA) was obtained from Merck and purified via vacuum distillation. N,N-Dimethylformamide (DMF) and N-methyl-2-pyrrolidone (NMP) from Merck were dried via vacuum distillation over CaH₂. Tetrahydrofuran (THF) was supplied from Aldrich and dried by distillation over sodium wire. The nanoclay used in this study was an organically modified clay with commercial name Cloisite 30B manufactured by Southern Clay Products (USA). Mild steel plates (15 × 3 cm) as working electrode were purchased from local market. They were pre-treated by mechanical polishing using sandpaper (P400) and rinsed with acetone and water, and then dried under ambient condition.

Synthesis of hydroxyl-terminated polyurethane-prepolymer containing tertiary amine groups (HPU)

In a three-necked polymerization reactor equipped with a mechanical stirrer, condenser, N₂ inlet and dropping funnel, were placed PTMEG (70.00 g, 0.07 mol) and IPDI (31.12 g, 0.14 mol). The reaction temperature was increased to 95 °C and continued until the free NCO content of the product reached to the calculated theoretical value. The reaction mixture was diluted with addition of NMP (45 ml) and cooled down to 30 °C. Then, TEA (25.9 g, 0.14 mol) dissolved in THF (20 ml) was charged into the dropping funnel and added into the reactor slowly. The stirring of reaction mixture at 30 °C was continued until the NCO peak (2270 cm⁻¹) disappeared totally at the FTIR spectrum of sample taken from the reaction kettle every 0.5 h. The product was transferred to a Teflon-coated mold and kept under vacuum at 80 °C until all solvents were evaporated.

Synthesis of blocked isocyanate curing agent containing tertiary amine group (BIC)

BIC was prepared according to the general procedure reported in our previous publication, except that trimethylolpropane was replaced with DEA.³⁶ Briefly, a three-necked round-bottomed flask (250 ml) equipped with condenser, dropping funnel, and N₂ inlet and outlet and magnetic stirrer was charged with IPDI (26.7 g, 0.12 mol) and DMF (20 g). Then a solution of DEA (7.15 g, 0.06 mol) in DMF (16 g) was dropped into the reaction mixture through the dropping funnel. The reaction temperature was increased slowly to 70 °C and the reaction was continued for 2 h. The flask was cooled down to room temperature and charged by dropwise addition of NMA (12.86 g, 0.12 mol) dissolved in DMF (10 g). The reaction temperature was raised to 45 °C and continued until complete disappearance of the NCO peak at 2270 cm⁻¹ in FTIR spectrum of the sample collected from reaction mixture. Then the product was freed from the solvent by casting into a Teflon-coated mold and placing in a vacuum oven at 70 °C.

Synthesis of amine functional Cloisite 30B (AMC)

Chemical modification of Cloisite 30B was performed according to the procedure reported by Wan.³⁷ In a 100 ml one-necked round-bottomed flask equipped with condenser and magnetic stirrer was placed APS (0.81 g) ethanol (96%, 25 ml). Then Cloisite 30B (1 g) was added into the flask and functionalization reaction was performed through constant stirring and refluxing of reaction mixture at 70 °C for 6 h. The resulting product purified by successive washing and centrifuging cycles with ethanol/water solvents mixture. The precipitated product was dried at 90 °C for 24 h.

Preparation of neat polyurethane coatings (PU) by CEPD procedure

The CEPD bath was prepared via protonation of a mixture consisting equivalent amounts of HPU (21.2 g) and BIC (18.9 g) dissolved in DMF (20 g) with excess amount of acetic acid (2.50 g) at 50 °C for 10 h. The total weight of solution was adjusted to 400 g (10 wt% solid content) by addition of appropriate amount of distilled water (340.00 g). The pH and electrolytic conductivity of prepared solution was measured as 5.5 and 2200 μS cm⁻¹ at 26 °C, respectively. A rectangular aluminum plate (15 × 3 cm) as anode equal in size and shape to a mild steel plate cathode, were immersed in the prepared solution. The distance of electrodes was adjusted to 5 cm. Then the anode and the cathode were connected to the respective terminals of a DC power supply (0–250 V, 6 A from HUAYI ELECTRONICS Company, China). The electrodeposition process was examined at three different constant voltages (70 V, 100 V, 120 V) for fixed period of 3 min. To obtain final PU coating, the electrodeposited film was removed from bath, rinsed with distilled water and subjected to thermal curing procedure at 200 °C for 2 h.

Preparation of polyurethane-clay nanocomposite coatings (CPU) by CEPD procedure

For preparation of CEPD bath containing AMC nanoparticles suitable for preparation of nanocomposites, the ingredients described at previous section were mixed and protonation reaction was carried out for 9 h. Then different formulations were prepared by introduction of various amounts of AMC (0.5, 1, 3 and 5 wt% respect to total weights of HPU and BIC) to resin ingredients. The protonation reaction was further continued at same condition for 1 h. Subsequently, distilled water was added and the prepared mixture was exposed to ultrasonic irradiation at 70 W using ultrasonic probe (ke 76) for 1.5 h. The CEPD procedure under previously fixed optimum condition was performed for 3 min, using bath solution containing different wt% of AMC. Then electrodes were removed and CEPD procedure was continued for additional 2 min in CEPD bath containing no AMC additive. The rest of procedure, i.e. rinsing and thermal treatment of coated cathode was continued till the final CPU nanocomposites were prepared.

Instruments and analytical methods

NCO content of NCO-terminated polyurethane intermediate was specified following the procedure reported in ASTMD2572. FTIR spectra were recorded using a Bruker instrument (model IFS48). All spectra were obtained in air, with 16 scans at a resolution of 4 cm⁻¹ and spectral range of 500–4000 cm⁻¹. ¹HNMR spectrum was obtained by a Bruker AVANCE DPX400 MHz instrument with CDCl₃ as a solvent. pH and conductivity of CEPD solution were determined by MultiMeter (model AL15 from AQUALYTIC company). Particle size was determined by Dynamic Light Scattering instrument (model SEM-633 from SEMA Tech). Measurement of coatings thickness was done utilizing MEGA CHECK 20-st digital thickness-tester. Hardness of cured coatings was determined by a pendulum hardness tester (Elcometer, model 3034, England) on Persoz mode according to ISO1522. Investigation of adhesion strength of prepared coatings to mild steel substrate was accomplished by a cross-cut adhesion tester (Elcometer, England) according to ASTM D3359. Falling ball test (ASTM D2794) and mandrel bending test (ASTM D522) were also utilized for the evaluation of impact resistance and flexibility of the coatings, respectively. Thermogravimetric analysis (TGA) was performed using Mettler-Toledo instrument (model TGA/DSC 1, Switzerland) from 25 to 600 °C at a heating rate of 10 °C min⁻¹ under nitrogen atmosphere. The X-ray diffraction (XRD) was performed at room temperature on a X-ray diffractometer (Siemens model D 5000) using Ni-filtered Cu Ka radiation (35 kV, 25 mA) with scanning rate of 1.2° min⁻¹.

Dynamic mechanical analysis (DMA) was performed using a Triton instrument (model Tritec 2000, England) in torsion mode in temperature range of −100 to 300 °C, heating rate of 10 °C min⁻¹ and frequency of 1 Hz. The values of storage modulus (G′), loss modulus (G′′) and loss tangent (tan δ) versus temperature were recorded for each sample. The maximum temperature of tan δ curves was considered as glass transition temperature (T_g) of samples. Crosslink density (ν_c) of the coatings was computed from their storage modulus in the rubbery region by utilizing following equation derived from the statistical theory of rubber elasticity.³⁸

G′ = Φν_cRT,

where G′ was the storage shear modulus at the onset of the rubbery plateau region obtained from DMA curves, Φ as the front factor that was considered as 1 for ideal rubber, ν_c as the cross-link density, defined as the mole number of network chains per unit volume of the polymer, R as the gas constant, and T the absolute temperature at the initiation of the rubbery plateau region. For measuring of gel content of the coatings, samples were weighed accurately and extracted by THF in a Soxhlet extractor for 24 h. The insoluble part was dried at 70 °C and weighed. The gel content was determined using following equation:

Gel content% = (W_d/W_i) × 100,

where W_d was the weight of dried sample after extraction and W_i was the primary weight of the sample. The values reported were an average of three measurements.

Electrochemical Impedance Spectroscopy (EIS) was carried out at a three electrode cell with a platinum plate (1 cm²) as auxiliary electrode, Ag/AgCl (KCl sat.) as reference electrode and coated mild steel (1 cm²) as working electrode, immersed in aqueous solution of NaCl 3.5 wt%. EIS measurements were performed in the frequency range from 100 kHz down to 10 mHz at the open circuit potential (OCP). Assessment of the corrosion behavior of the electrodeposited coatings was done based on Nyquist and Bode plots recorded during different immersion time in NaCl 3.5 wt% solution. Also, salt spray test (ASTM B117) was performed and coatings resistance against corrosive media (NaCl 5 wt% fog at 35 °C) were visually examined at different exposure time.

Results and discussion

Synthesis and characterization of reactants

Preparation of cataphoretically applicable two-component, one-pack waterborne polyurethane system with ease of handling, simplicity of application on surfaces with complex morphology, very good physico-mechanical and anticorrosion performance was considered in this study. This system was comprised of HPU as a hydroxyl-terminated polyurethane precursor and BIC as a blocked isocyanate curing agent. To make these components water reducible and active for cataphoretically electrodeposition process, tertiary amine groups was inserted in their backbone. The synthetic route followed for the preparation of HPU is depicted in Scheme 1.

Scheme 1 Synthetic route for the preparation of HPU.

The reaction of PTMEG with an excess amount of IPDI led to the formation of NCO-terminated polyurethane prepolymer, which subsequently converted to hydroxyl functional material (HPU) via reaction with TEA. In this procedure, TEA was dropped into the reactor containing NCO-terminated prepolymer to provide suitable condition for the partial chain extension of two reactants. This method is favor for the preparation of HPU with asymmetrically grown (chain extended) structure and consequently better distribution of hydroxyl groups in a larger molecule that can lead to better flexibility of the final coating.

The structure of HPU was elucidated by FTIR, and ¹HNMR spectroscopy.

In FTIR spectrum of HPU (Fig. 1a) the broad peak centered at 3325 cm⁻¹ was attributed to the stretching vibration of urethane N–H and alcoholic O–H groups overlapped with each other. Characteristic peak of urethane carbonyl bond (NH–CO–O) was revealed at 1700 cm⁻¹ (mainly H-bonded). The peak appeared at 1539 cm⁻¹ was related to C–N stretching, combined with N–H out-of-plane bending of urethane groups. The stretching vibration of ether C–O groups in PTMEG appeared as a peak at 1103 cm⁻¹. Peaks correspond to stretching and bending vibrations of CH, CH₂, CH₃ groups were observed at 2931, 2850, 1461 and 1305 cm⁻¹. HPU structure was also studied by ¹HNMR spectroscopy (Fig. 1b). Terminal methyl and methylene groups of isophorone ring appeared at 0.94–1.2 ppm. The peaks observed at 1.63 ppm and 3.4 ppm were ascribed to central methylene groups and methylene groups attached to etheric oxygen of PTMEG moieties. Characteristic peak of methylene group attached to urethane oxygen atom originate from PTMEG segments was detected at 4.1 ppm. The peak of methylene group derived from TEA segment attached to urethane oxygen atom was appeared at 4.5 ppm. Vicinity of this methylene group to electron withdrawing nitrogen atom was the reason for shift to higher ppm value. The peak of methylene groups attached to nitrogen atom of TEA was also detected at 2.9 and 3 ppm. Average degree of polymerization (DP_n) and number average molecular weight (M_n) of HPU was estimated using Carother's equation (Table 1). Hydroxyl equivalent weight of HPU was also calculated and result was presented in Table 1. The calculated DP_n value (3) was in good agreement with what estimated by considering the area of f to g peaks at ¹HNMR spectrum of HPU (2.7). Therefore, no gel permeation chromatography experiment was considered in this work for estimation of reactants molecular weight. The calculation of block ratio of reactants was performed based on what extracted from Carother's equation.

Fig. 1 (a) FTIR and (b) ¹HNMR spectrums of HPU, FTIR spectra of (c) BIC and (d) cured coating.

Table 1 Characteristics of HPU and BIC

Sample	Mn^a (g mol^−1)	Average equivalent weight^b
a Calculated Mn based on Carothers equation, DPn = (1 + r)/(1 − r), r = 0.5, Mn = DPn × M0.b Average OH equivalent weight = (n1M1 + n2M2 + n3M3)/[4/3(n1 − n2 − 2n3)], where n1, n2 and n3, and M1, M2 and M3, are the molar ratios and molecular weights of PTMEG, IPDI and TEA, respectively. NCO equivalent weight = 4200/NCO%, NCO% = 2 × 42 × 100/[119.2 + (2 × 222.3) + (2 × 107.2)].
HPU	4188	435
BIC	778	389

---

The second component, tertiary amine functional blocked isocyanate (BIC) was prepared through the chemical reactions depicted in Scheme 2.

Scheme 2 The synthetic route for the preparation of BIC.

At first, IPDI capped DEA intermediate was synthesized and then free isocyanate groups were blocked by reaction with NMA. Preparation of similar blocked isocyanates using NMA as blocking agent was reported previously by us and others. Lower deblocking temperature was expected when NMA used instead of phenol as common blocking agent. Formation of an intermolecular four-centered transition state between the tertiary nitrogen in urea moiety and NMA was responsible for the lower activation energy of deblocking reaction.^36,39 Also, NMA is expected to be less corrosive than common phenolic compounds, therefore, crosslinker based on this blocking agent is more suitable for making heat-curable PU coatings for metallic surface. The prepared BIC was characterized by FTIR spectroscopy (Fig. 1c). In this spectrum stretching vibration of urethane N–H was observed as a broad peak centered at 3321 cm⁻¹. The peaks appeared at 1711 and 1666 cm⁻¹ were attributed to urethane and urea carbonyl bonds (NH–CO–O, NH–CO–N), respectively. The peak observed at 1520 cm⁻¹ was related to C–N stretching, combined with N–H out-of-plane bending of urethane linkage. Characteristic peak of C–H and C=C bonds related to aniline aromatic ring were seen at 3057 and 1600 cm⁻¹, respectively. Also, peaks related to stretching and bending vibrations of CH, CH₂, CH₃ groups were observed at 2950, 2915, 1460 and 1384 cm⁻¹. The calculated NCO equivalent weight of BIC was also presented in Table 1.

AMC was applied as a modified clay additive to increase the barrier property of the prepared electrodeposited polyurethane coatings. It is well-known that negative carboxylate groups are present in clay structure; therefore, this additive is not suitable for cathodic electrodeposition procedure, since this material tends to migrate to anode instead of cathode during electrodeposition procedure. Amine groups were decorated on the clay structure to provide necessary condition for cataphoretically electrodeposition of this material. In fact, the positive electric charge developed in the clay structure after protonation of embedded amine groups in acidic medium is responsible for migration toward cathode. As it is shown in Scheme 3 AMC was simply prepared via reaction of Cloisite 30B with an amine functional silane coupling agent (APS).

Scheme 3 Synthetic route followed for the preparation of AMC.

The structure of modified clay was characterized by FTIR, XRD and TGA methods and data was compared with clay before chemical modification (Fig. 2). In FTIR spectrum of AMC, peaks observed at 3631 and 3397 cm⁻¹ were related to stretching vibrations of the hydroxyl groups (O–H) that were bonded to the Al, Mg, Si and the adsorbed water, respectively. In plane stretching vibration and out of plane bending vibration of Si–O bonds were appeared as peaks at 1047 cm⁻¹ and 523 cm⁻¹, respectively. The observed peaks at 917 and 800 cm⁻¹ were due to the bending vibrations of Al(Al)–OH and Mg(Mg)–OH, respectively.³⁷ Stretching vibration of C–N bonds arose from organic modifier of clay appeared at 1654 cm⁻¹. As well, stretching vibrations of CH₂, CH₃ groups were detected as two peaks at 2925 cm⁻¹ and 2851 cm⁻¹, and related bending vibrations were appeared at1469 cm⁻¹. After modification with APS, the intensity of Si–O peak increased. This was due to the reaction with silanol groups present on the broken edges of the silicate layers and more importantly formation of polysiloxane oligomers through sol–gel reaction of siloxane groups of APS (Fig. 2a).

Fig. 2 (a) FTIR spectra, (b) TGA thermograms, and (c) XRD spectra of Cloisite 30B and AMC.

Evaluation thermal stability of AMC and comparison with Colisite 30B revealed extra confirmation for the possible formation of polysiloxane oligomers, since thermal stability of AMC was enhanced significantly. Coverage of the clay surface and reduction of heat transfer into internal covered layer of clay was the proposed reason for this observation (Fig. 2b).³⁷

The XRD patterns of Cloisite 30B before and after modification reaction are illustrated in Fig. 2c. The characteristic diffraction peak at 4.9° related to the plane reflection of the Cloisite 30B slightly shifted to smaller angle (4.7°) after modification reaction. Employing Bragg's equation revealed that the basal layers spacing increased from 1.82 nm to 1.90 nm after modification reaction. The increase in d-spacing of the silicate layers was attributed to the intercalation of APS into the clay intergalleries as the result of reaction with either Si–OH groups present on edge of silicate layer and O–H groups of, bis(2-hydoxyethyl) methyl hydrogenated tallow ammonium chloride used as organic modifier in Cloisite 30B.³⁷

It is worth mentioning that despite of few reports regarding anodic electrodeposition of clay, no corresponding data is available for the CEPD of clay. Therefore, this work can open a new perspective for the modification of commercially available coatings designed for CEPD process. The results regarding enhanced properties of polyurethane coating system synthesized in the present study are described in the following sections.

Preparation and characterization of PU coating

Particle size of the dispersed components in electrodeposition bath can drastically influence on characteristics of the system such as stability and shelf life of dispersion, the kinetic of film growth, electrophoretic velocity of charged particles and dry film thickness.^40,41 Therefore, the particle sizes of aqueous dispersion of protonated HPU, BIC and a mixture composed of the equivalent amounts of HPU and BIC at solid content of 10 wt% were measured (Fig. 3). The prepared HPU dispersion showed lower and narrower size distribution than BIC component, which may resulted from higher charge density of HPU. It is worth mentioning that, film grows in a porous like manner from dispersion containing too small particles which is attributed to the higher rate of hydrogen gas evolution than the rate of electrocoagulation of particle on the surface. Meanwhile, for systems consisting large particles, the rate of gas evolution is much lower than the rate of coagulated particles deposition. Therefore, the gas bubbles can be trapped within uncured films. This may again leads to formation of porous film after thermal curing procedure. The dispersion made from mixing and protonating of HPU and BIC showed particle size of about 100 nm and moderate particle size distribution. Therefore similar electrophoretic velocity during electrodeposition procedure was expected for these two components. As well, acceptable quality of electrodeposited film was expected for this mixture, since the particle size of the mixture components was comparable with some of related successful formulations.^17,42,43

Fig. 3 Particle size distribution curves of HPU, BIC and their mixture.

The HPU/BIC mixture was subjected to CEPD process via applying three different DC voltages (70, 100 and 120 V) for 3 min and constant temperature of 26 °C. The electrodeposited films were then subjected to thermal curing procedure at fixed temperature of 200 °C for 1 h. The thickness of final cured coatings was measured and considered as a parameter for optimization of electrodeposition step. Increasing the applied voltage from 70 to 100 V led to increased thickness of cured coatings from 21 ± 2 μm to 28 ± 2 μm. However, at voltage of 120 V the rate of gas bubble generation at cathode was much higher than coagulation of electrodeposited particles. This phenomenon was prevented complete coverage of electrode surface and finally film rupture was occurred. Therefore, the DC voltage of 100 V was chosen as optimum voltage for the rest of experiments. To find information regarding chemical structure of the prepared material and suitability of thermal curing condition, the coating layer obtained at optimum condition was scratched from the surface of cathode electrode and its FTIR spectrum was recorded (Fig. 1d). FTIR spectrum of this polymer showed characteristic peak of urethane carbonyl groups in the range of 1720–1742 cm⁻¹. The urea linkage originate from BIC was detected as a peak at 1656 cm⁻¹. Urethane N–H stretching bond was appeared at 3400 cm⁻¹. Combination of C–N stretching and N–H out of plan bending was appeared as a peak at 1562 cm⁻¹. Stretching vibration of ether (C–O–C) groups from PTMEG moieties was also detected as a peak at 1115 cm⁻¹. There was no sign of free isocyanate peak at 2270 cm⁻¹ that confirmed the complete reaction of isocyanate groups of curing agent. Then the prepared coating was subjected to different assessments.

First of all, the quality of electrodeposited and cured film was visually examined by optical microscopy (Fig. 4). Film with uniform coverage of substrate without detectable pinholes was obtained which confirmed suitable condition applied for the electrodeposition and curing procedures. The high value of gel content (>80%) and hardness (257 s) of resulting coating measured by pendulum hardness test confirmed appropriate progress of crosslinking and curing reaction. The impact resistance of coating for both direct and reverse impact test was also examined. The coated mild steel panel showed no sign of damage to the film on either case (0.34 kg m). Evaluating of the coating flexibility through bending on mandrels with different diameter was also showed high flexibility with no cracks or rupture when mandrel of 5 mm diameter was used (Fig. 5a). High intermolecular attraction of polar urethane and urea groups and co-presence of flexible ether linkages arose from PTMEG moieties were the reasons for the formation of such a tough coating material.⁴⁴ Adhesion of PU coating to mild steel was evaluated by cross-cut test (Fig. 5b). The edges of the cuts were thoroughly smooth and none of the blocks of coating on the grid squares were detached while removing adhered tape, as a result the coating showed excellent adhesion of grade 5B to metal surface. This feature was attributed to the strong interaction of polar urethane, urea and ether bonds present in the backbone of polyurethane coating with polar hydroxyl groups available on the metal surface.⁴⁴

Fig. 4 The coated plate at different magnification (a) normal, (b) ×4 and (c) ×10.

Fig. 5 (a) Image of bended coated plate, (b) optical microscopic image (×4) of PU coating after cross-cut test.

Preparation and characterization of CPU nanocomposites

After optimization of general properties of polyurethane coating, and knowing the importance of barrier mechanism for improving the corrosion protection of organic polymeric coating materials, direct introduction of clay additive during electrodeposition procedure was examined. For this purpose, clay nanoparticles chemically decorated with amine groups (AMC) was prepared and mixed with HPU and BIC components. Different amounts of AMC (0.5, 1, 3 and 5 wt%) was considered respect to total amount of HPU and BIC materials. The resulting electrodeposition baths were subjected to CEPD at 100 V for 3 min. The behavior of system containing AMC was different with similar system without added AMC. While applying voltage to system with no AMC, the passing current gradually decreased and reached to almost zero ampere. But for those systems containing AMC, an initial decrease in passing current was detected, then the current increased slightly accompanied with warming of the electrodeposition bath up to 40 °C. In contrast to neat system, the current never reached to zero ampere. Comparing the quality of electrodeposited films, it was revealed that formation of a continuous and uniform film was only possible at the low examined concentration of AMC (0.5 and 1 wt%), therefore only these two compositions were subjected to further evaluations. In fact, at high concentration of AMC, agglomeration of clay particles was occurred and subsequently larger particles were deposited along with PU matrix. This phenomenon was led to formation of porous films on the surface of cathode. Because of incomplete coverage of cathode, the redox reactions and generation of H₂ gas and OH⁻ ions were continued all over the electrodepositing period. Additionally, increasing the solution temperature and subsequently increased mobility of the PU moieties perturbed the optimum condition for coordinated electrodeposition of these ingredients. It is important to note that even at the lowest concentration of AMC, the passing current was not reached to zero which was attributed to the presence of pinholes in the deposited film. To solve this issue, the coated electrodes were again put into bath containing matrix ingredients without added AMC and the voltage of 100 V was applied for additional 2 min. The coated cathodes were again subjected to thermal curing procedure and uniformity of coated layer was visually examined (Fig. 6). Inspection of recorded pictures confirmed entire coverage of substrates for both formulations. Meanwhile, the thickness of coatings obtained for these two formulations CPU0.5 (90 ± 6) and CPU1 (81 ± 13) were measured. Film with better uniformity at higher thickness was obtained for lower concentration of AMC.

Fig. 6 Images of electrodeposited coatings, CPU0.5: (a) normal, (b) ×4 magnification,CPU1: (c) normal, and (d) ×4 magnification.

The CPU0.5 and CPU1 samples were subjected to XRD analysis. The X-ray diffractograms for these samples and AMC are shown in Fig. 7a.

Fig. 7 (a) XRD diffractograms of AMC, CPU0.5 and CPU1; (b) TGA thermograms of PU, CPU0.5 and CPU1; variation of (c) storage modulus and (d) tan δ vs. temperature for PU and CPU0.5.

As it was previously described, the characteristic diffraction peak of AMC was clearly detected at 4.7°; however no related diffraction peak was left in nanocomposites samples. Also, no obvious difference was detected at two levels of added AMC. This can be attributed to complete exfoliation of AMC plates in the matrix of polyurethane coating.^32,45

Influences of some factors are suggested for satisfied dispersion and exfoliation of AMC in polymeric matrix. Water was served as an ideal dispersing medium for protonated cationic form of HPU, BIC and AMC. Also, the mass content of matrix components (HPU and BIC) was much higher than AMC; therefore, they could prevent coagulation of positively charged AMC particles in strong electric field generated during rapid electrodeposition process. Thermal stability of prepared polyurethane coating and corresponding nanocomposites was studied by TGA. As it is shown in Fig. 7b, two decomposition stages were detected for coatings. The first one started at about 250 °C was related to thermal dissociation of urethane, and urea bonds present in hard segment of polyurethane matrix. The second stage of thermal decomposition was associated to the degradation of linkages with higher thermal stability such as etheric bonds originate from polyol structure.⁴⁶ After incorporation of AMC, the decomposition temperature of coatings especially at second stage of decomposition was slightly shifted to higher value. This partial improvement in the thermal stability was related to the clay additive which acted as a thermal insulator for heat transfer to polymeric matrix. AMC was also acted as a mass transport barrier for releasing of volatile products formed during thermal decomposition of polymeric chain segments.^32,47 However, due to low percentage of added AMC, the improvement of thermal stability was not significant.

Thermo-mechanical property of coatings with and without added AMC was analyzed by DMA. The experiments were performed in torsion mode. The variation in storage modulus (G′), and loss tangent (tan δ) as a function of temperature was monitored (Fig. 7c and d). Obviously, the PU coating showed a tan δ peak at 74 °C, corresponding to the T_g of soft polyol segment. Compared to neat polyol T_g (−88 °C),⁴⁸ the recorded increase of T_g confirmed the restriction for chain mobility due to formation of a cross linked network upon thermal treatment. However tan δ peak of the nanocomposite appeared at 10 °C lower temperature compared to PU coating. This result could be assigned to the reduction of crosslink density in nanocomposite due to the presence of nanoparticles.^49–51 In fact, nanoparticles acted as physical barrier and reduced accessibility of HPU and BIC components for reaction with each other and formation of crosslinked network. This concept was further confirmed by quantifying the crosslinked density of samples using pre-mentioned equation derived from theory of rubber elasticity. The calculated crosslink density was 13 (mol m⁻³ × 10⁻⁵) for PU coating and 12 (mol m⁻³ × 10⁻⁵) for CPU0.5 sample. Close inspection of DMA curves was also revealed that addition of clay nanoparticles into the coating matrix caused a significant enhancement in storage modulus below the T_g attributed to high modulus of inorganic nanoparticles. However, after passing from T_g, the modulus of nanocomposite sample reduced to lower amount in comparison to neat sample related to lower crosslink density of resulting network.

Anticorrosion property of coatings

Electrochemical assessment. EIS analysis was performed in order to find quantitative information regarding corrosion resistance of coating and nanocomposites. The coated plates were immersed in corrosive media, and Nyquist and Bode plots were recorded. In Nyquist plot the diameter of the observed semicircle is a measure for resistance of coatings against electrolyte diffusion which is named pore resistance (R_p). For acquiring electrochemical parameters of the coatings notably R_p in this study, an equivalent circuit should be considered as a model. Randle's model⁵² consisting three components as electrolyte resistance (R_s), R_p and constant phase element (Q) is an appropriate equivalent circuit for simulating the Nyquist plots with one semicircle.

The Nyquist plots recorded for samples immersed in corrosive media (NaCl 3.5%) are shown in Fig. 8. After passing 2 h from immersion in corrosive media and simulating the recorded data, the corrosion resistance for about 8 GΩ and 5 GΩ were obtained for CPU0.5 and CPU1 samples, respectively. Under the same condition, much lower R_p value (1.2 GΩ) was obtained for neat PU coating. The superior property of CPU0.5 and CPU1 samples was related to lower permeability of these coatings for corrosive species due to the barrier effect of dispersed nanoclay particles. After elapsing 4 days from immersion in corrosive media, all of the samples showed decreased amount of R_p value due to possible diffusion of electrolytes into the coatings. By extending the immersion time to 7 days, two semicircles were appeared in the Nyquist plot of neat PU coated sample which corresponded to the second time constant and double electrical layer capacitance.

Fig. 8 Nyquist plots of coatings after passing (a) 2 h, (b) 4 days, (c) 7 days, (d) 10 days, (e) 14 days and (f) 17 days from immersion in NaCl 3.5% solution.

This phenomenon was related to the diffused electrolytes reached to the substrate surface and starting of electrochemical corrosion reactions at the coating–metal interface. However, at the same condition the nanocomposite samples preserved their high corrosion resistance (7 × 10⁸ Ω for CPU0.5 and 7.0 × 10⁷ Ω for CPU1), due to the better ability of coatings for prevention of corrosive species diffusion. After passing 14 days from immersion of coated samples in corrosive media, a considerable reduction in R_p of CPU1 was detected, but no obvious variation in corrosion resistance of CPU0.5 sample was observed. Eventually, with lengthening of the immersion time to 17 days and the presence of the second semicircle in Nyquist plot of CPU0.5 and CPU1 samples, it was revealed that the corrosion reaction started on the surface of substrate. The precise values of obtained pore resistance by Nyquist plots with one semicircle were tabulated in Table 2. The recorded data was in agreement with the few studies regarding incorporation of nanoclay into the coating formulation by electrophoretic deposition. For example De Riccardis et al.⁵³ were studied the incorporation of montmorillonite in epoxy resin by electrophoretic deposition method. For this system, the R_p of the neat epoxy resin was improved from 344.70 kΩ cm² to 539.2 6 kΩ cm² after incorporation of 5 wt% clay (10 min deposition time).

Table 2 The pore resistance of the samples obtained by fitting the EIS data in Randle's model and Bode plots

Immersion time in NaCl 3.5%	PU		CPU0.5		CPU1
	Nyquist	Bode	Nyquist	Bode	Nyquist	Bode
2 h	1.2 × 10^9	1.1 × 10^9	8.0 × 10^9	7.4 × 10^9	5.1 × 10^9	6 × 10^9
4 days	8.1 × 10^7	7.6 × 10^7	8.0 × 10^8	8.0 × 10^8	7.3 × 10^7	7.3 × 10^7
7 days	—	7.0 × 10^3	7.1 × 10^8	6.8 × 10^8	6.2 × 10^7	6.3 × 10^7
10 days	—	—	7.0 × 10^8	6.5 × 10^8	7.0 × 10^7	6.9 × 10^7
14 days	—	—	7.0 × 10^8	6.5 × 10^8	1.8 × 10^6	2.1 × 10^6
17 days	—	—	—	1.4 × 10^5	—	1.9 × 10^4

---

The study of bode plots which display the logarithm of the impedance modulus, |Z|, and phase angle, φ, versus the logarithm of the frequency is a versatile and simple method for investigation of anticorrosion behavior of coatings, especially when proper modeling of Nyquist plots is not possible. In this method, impedance at low frequency (10 mHz) represents R_p of coatings.^54,55 As it was shown in Fig. 9a after passing 2 h from immersion in NaCl solution, higher resistance was recorded for nanocomposite samples (about 10¹⁰ Ω) than neat PU coating (10⁹ Ω) at frequency of 10 mHz. By extending immersion time to 4 days reduction in R_p was occurred due to partial diffusion of electrolyte into the coatings (Fig. 9b).

Fig. 9 Bode plots of coatings after passing (a) 2 h, (b) 4 days, (c) 7 days, (d) 10 days, (e) 14 days and (f) 17 days passed from immersion in NaCl 3.5% solution.

The complete penetration of electrolyte through the neat PU coating was occurred after 7 days passed from immersion in corrosive media and failure of anticorrosion property of coating was realized due to diminishing the R_p value below 10⁶ Ω (Fig. 9c).⁵⁶ However, the R_p values remained above than 10⁶ Ω for nanocomposite coatings until 14 days passed from exposing to corrosive media, confirmed better barrier performance of nanocomposites and longer tortuosity of the diffusion pathway for electrolyte species (Fig. 9e). Eventually after 17 days, the corrosive species found their pathway to substrate surface resulted in reduction of coatings resistance to values lower than 10⁶ Ω (Fig. 9f). The pore resistance values of coatings derived from Bode plots were also collected in Table 2.

Breakpoint frequency is appeared when a system undergoes a capacitor to resistor transition and is determined by the frequency at which the phase angle between the perturbation voltage and the response current is 45°.^57–60 According to the following equation, there is a direct correlation between breakpoint frequency (f_b) and delamination area (D) of the coatings.

f_b = D/2περ

f_b is also influenced by the dielectric permittivity (ε) and the specific resistance (ρ) of the coating. However, previous studies revealed that ε of a coating increases no more than 10-fold during the immersion period.^59,61 Therefore, the increase of f_b can be mainly attributed to decrease of ρ and/or increase of D. From Bode plots it was found that after 2 h immersion in NaCl 3.5% solution, the nanocomposites with the f_b of about 0.15 Hz showed more capacitive behavior than neat PU coating with the f_b of 0.5 Hz. For all samples, lengthening of immersion time to 4 days caused an increase in f_b values as a result of increased amounts of diffused electrolyte into the coatings and consequently, growing the delamination area of coatings. Therefore, current could pass through the films with more ease and resistivity behavior was detected at higher frequency. After 7 days, no f_b was detected in the Bode plot of neat PU coating, however for nanocomposites the f_b as detected and remained at the appropriate values. At day 17, due to complete saturation of coatings with electrolyte species, the nanocomposite coatings were also acted as resistor and no f_b was detected for them in entire frequency range. It is worth mentioning that all of electrochemical experiments confirmed better corrosion inhibition behavior of CPU0.5 than other samples. The f_b amounts obtained by Bode plots are reported in Table 3.

Table 3 The breakpoint frequency of the samples obtained by Bode plots

Immersion time in NaCl 3.5%	PU	CPU0.5	CPU1
2 h	0.5	0.15	0.15
4 days	3.16	1	5
7 days	—	2.5	10
10 days	—	2.5	10
14 days	—	3.31	316
17 days	—	—	—

---

Salt spray assessment. Salt spray test was carried out as a qualitative method for visual inspection of coatings resistance against corrosive environment. The images of coated plates before and after subjecting to this test at different exposure time are presented in Fig. 10. After 4 days exposure, some blisters were observed and corrosion process was extended under the scratched area in neat PU coating, but the nanocomposite samples were free from blisters. Also little amounts of rusts were formed along the scratched regions. The area of rusted regions was smaller for CPU0.5 than other samples. By extending the exposure time, the corrosion process was progressed in all samples. At day 12, the better performance of CPU0.5 than other samples was clearly evidenced by lower amount of damaged area (lower amounts of blisters and rusted surface). After 17 days, the corrosion was progressively extended under coatings; still deterioration of coating performance was lower for CPU0.5 sample.

Fig. 10 Images of coated samples exposed to NaCl 5 wt% fog at different time intervals.

The recorded experimental data regarding resistance of the designed coatings against corrosive media, confirmed the usefulness of added modified clay nanoparticles for improving the practical performance of coatings. Our findings was also in accordance to the works related to incorporation of other types of nanoparticles into EPD formulations. For example Zhu et al.⁶² were observed that by addition of PU decorated SiO₂ nanoparticles into the epoxy resin, the salt-fog tolerance/days of neat resin was increased from 30 to 50. Ranjbar et al.²⁴ were used nano-zinc oxide as an additive in cathodically electrodeposited polyurethane based films. It was found that the film resistance of the nano-ZnO filled polyurethane-based waterborne coatings was much higher than that of the neat coatings. The nano-ZnO filled films maintain their protective property even after 120 days of immersion. Also based of EIS data and bode plots, a corrosion resistance of 2 orders of magnitude higher than that of the neat films was observed for nanocomposite.

Conclusion

A waterborne two-component one-pack cationic polyurethane system composed of a hydroxyl terminated urethane prepolymer and a block isocyanate curing agent was synthesized and subjected to CEPD process under optimized condition. The PU coating showed excellent adhesion of 5B to mild steel surface, high impact resistance equal to 0.343 (kg m), high hardness, excellent flexibility and good corrosion resistance up to 7 days exposure to corrosive media. For the first time, a surface modified clay was co-deposited on cathode along with PU components during CEPD process. The nanocomposite coatings at two different levels of nanoclay additive were subjected to different characterization methods. The resistance of these clay-modified coatings against corrosive media was significantly improved as determined by EIS and salt spray studies. For the best formulation of clay incorporated coating, the failure time of coating was increased up to 17 days. Longer tortuosity of the diffusion pathway for electrolyte species and therefore, better barrier property of coatings containing well dispersed clay nanoparticles was the reason for improved anticorrosive behavior of these coatings.

To scale up the process on larger area, two sets of parameters; including those related to the suspension such as particle size, dielectric constant, conductivity, zeta potential and viscosity of suspension, as well as those related to the process including the physical parameters such as the electrical nature of the electrodes and the electrical conditions (voltage/intensity relationship, deposition time), should be optimized. These tasks will be considered in future research.

References

1. A. Forsgren, Corrosion Control Through Organic Coatings, CRC Press Book, 2006, https://www.crcpress.com/Corrosion-Control-Through-Organic-Coatings/Forsgren/9780849372780, accessed November 16, 2015.
2. E. McCafferty, Introduction to Corrosion Science, Springer, 2010,  DOI:10.1017/cbo9781107415324.004.
3. M. F. Montemor, Functional and smart coatings for corrosion protection: a review of recent advances, Surf. Coat. Technol., 2014, 258, 17–37,  DOI:10.1016/j.surfcoat.2014.06.031.
4. P. A. Sørensen, S. Kiil, K. Dam-Johansen and C. E. Weinell, Anticorrosive coatings: a review, J. Coat. Technol. Res., 2009, 6, 135–176,  DOI:10.1007/s11998-008-9144-2.
5. A. Overbeek, Polymer heterogeneity in waterborne coatings, J. Coat. Technol. Res., 2010, 7, 1–21,  DOI:10.1007/s11998-009-9201-5.
6. T. F. L. McKenna, B. Charleux, E. Bourgeat-Lami, F. D'Agosto and M. Lansalot, Novel technologies and chemistries for waterborne coatings, J. Coat. Technol. Res., 2014, 11, 131–141,  DOI:10.1007/s11998-013-9554-7.
7. L. Besra and M. Liu, A review on fundamentals and applications of electrophoretic deposition (EPD), Prog. Mater. Sci., 2007, 52, 1–61,  DOI:10.1016/j.pmatsci.2006.07.001.
8. A. R. Boccaccini and J. H. Dickerson, Electrophoretic deposition: fundamentals and applications, J. Phys. Chem. B, 2013, 117, 1501,  DOI:10.1021/jp211212y.
9. M. S. Ata, Y. Liu and I. Zhitomirsky, A review of new methods of surface chemical modification, dispersion and electrophoretic deposition of metal oxide particles, RSC Adv., 2014, 4, 22716,  10.1039/c4ra02218a.
10. Electrophoretic Deposition: Fundamentals and Applications V, ed. A. R. Boccaccini, J. H. Dickerson, B. Ferrari, O. Van der Biest and T. Uchikoshi, TRANSTECH PUBLICATIONS INC., Hernstein, Austria, Category: Selected, peer reviewed papers from the 5th International Conference on Electrophoretic Deposition: Fundamentals and Applications, October 5–10, 2014, Edition: softcover, Key Engineering Materials, 2015, vol. 654, ISBN-13: 978-3-03835-497-0, p. 300.
11. S. Seuss and A. R. Boccaccini, Electrophoretic deposition of biological macromolecules, drugs, and cells, Biomacromolecules, 2013, 14, 3355–3369,  DOI:10.1021/bm401021b.
12. M. Bučko, V. Mišković-Stanković, J. Rogan and J. B. Bajat, The protective properties of epoxy coating electrodeposited on Zn–Mn alloy substrate, Prog. Org. Coat., 2015, 79, 8–16,  DOI:10.1016/j.porgcoat.2014.10.010.
13. C. J. Patel and A. Dighe, Novel isocyanate-free self-curable cathodically depositable epoxy coatings: Influence of epoxy groups on coating properties, Prog. Org. Coat., 2007, 60, 219–223,  DOI:10.1016/j.porgcoat.2007.07.015.
14. T. Wang, S. Qi, B. Ren and Z. Tong, Preparation and surface characteristics of low-temperature curing fluorinated cathodic electrodeposition coating, Prog. Org. Coat., 2007, 60, 132–139,  DOI:10.1016/j.porgcoat.2007.07.009.
15. Y. B. Kim, H. K. Kim and J. W. Hong, UV-curable methacrylic epoxy dispersions for cationic electrodeposition coating, J. Appl. Polym. Sci., 2006, 102, 5566–5570,  DOI:10.1002/app.24766.
16. J. Li, L. Ecco, G. Delmas, N. Whitehouse, P. Collins and F. Deflorian, et al., In Situ AFM and EIS Study of Waterborne Acrylic Latex Coatings for Corrosion Protection of Carbon Steel, J. Electrochem. Soc., 2014, 162, C55–C63,  DOI:10.1149/2.0851501jes.
17. Y. B. Kim, H. K. Kim, J. K. Yoo and J. W. Hong, UV-curable polyurethane dispersion for cationic electrodeposition coating, Surf. Coat. Technol., 2002, 157, 40–46.
18. R. Liu, X. Li, F. An, S. Zhang and X. Liu, A photosensitive copolymer for UV-curable electrodeposition coatings, J. Wuhan Univ. Technol., Mater. Sci. Ed., 2011, 26, 1098–1102,  DOI:10.1007/s11595-011-0370-6.
19. R. Gharibi, M. Yousefi and H. Yeganeh, Synthesis, characterization and assessment of poly(urethane-co-pyrrole)s derived from castor oil as anticorrosion coatings for stainless steel, Prog. Org. Coat., 2013, 76, 1454–1464,  DOI:10.1016/j.porgcoat.2013.05.035.
20. X. Shi, T. A. Nguyen, Z. Suo, Y. Liu and R. Avci, Effect of nanoparticles on the anticorrosion and mechanical properties of epoxy coating, Surf. Coat. Technol., 2009, 204, 237–245,  DOI:10.1016/j.surfcoat.2009.06.048.
21. G. Christopher, M. Anbu Kulandainathan and G. Harichandran, Comparative study of effect of corrosion on mild steel with waterborne polyurethane dispersion containing graphene oxide versus carbon black nanocomposites, Prog. Org. Coat., 2015, 89, 199–211,  DOI:10.1016/j.porgcoat.2015.09.022.
22. B. P. Singh, S. Nayak, K. K. Nanda, B. K. Jena, S. Bhattacharjee and L. Besra, The production of a corrosion resistant graphene reinforced composite coating on copper by electrophoretic deposition, Carbon, 2013, 61, 47–56,  DOI:10.1016/j.carbon.2013.04.063.
23. B. Long, C.-A. Wang, W. Lin, Y. Huang and J. Sun, Polyacrylamide-clay nacre-like nanocomposites prepared by electrophoretic deposition, Compos. Sci. Technol., 2007, 67, 2770–2774,  DOI:10.1016/j.compscitech.2007.02.007.
24. M. Rashvand and Z. Ranjbar, Effect of nano-ZnO particles on the corrosion resistance of polyurethane-based waterborne coatings immersed in sodium chloride solution via EIS technique, Prog. Org. Coat., 2013, 76, 1413–1417,  DOI:10.1016/j.porgcoat.2013.04.013.
25. Q. Chen, U. P. de Larraya, N. Garmendia, M. Lasheras-Zubiate, L. Cordero-Arias and S. Virtanen, et al., Electrophoretic deposition of cellulose nanocrystals (CNs) and CNs/alginate nanocomposite coatings and free standing membranes, Colloids Surf., B, 2014, 118, 41–48,  DOI:10.1016/j.colsurfb.2014.03.022.
26. I. Zhitomirsky, Electrophoretic deposition of organic–inorganic nanocomposites, J. Mater. Sci., 2006, 41, 8186–8195,  DOI:10.1007/s10853-006-0994-7.
27. F. Pishbin, V. Mouriño, S. Flor, S. Kreppel, V. Salih and M. P. Ryan, et al., Electrophoretic deposition of gentamicin-loaded bioactive glass/chitosan composite coatings for orthopaedic implants, ACS Appl. Mater. Interfaces, 2014, 6, 8796–8806,  DOI:10.1021/am5014166.
28. J. Yeh and K. Chang, Polymer/layered silicate nanocomposite anticorrosive coatings, J. Ind. Eng. Chem., 2008, 14, 275–291,  DOI:10.1016/j.jiec.2008.01.011.
29. T. Thi Xuan Hang, T. A. Truc, T. H. Nam, V. K. Oanh, J.-B. Jorcin and N. Pébère, Corrosion protection of carbon steel by an epoxy resin containing organically modified clay, Surf. Coat. Technol., 2007, 201, 7408–7415,  DOI:10.1016/j.surfcoat.2007.02.009.
30. Y. W. Chen-Yang, C. H. Yang, G. J. Li and Y. K. Li, Thermal and Anticorrosive Properties of Polyurethane/Clay Nanocomposites, Polym. Res. J., 2004, 275–283.
31. M. Heidarian, M. R. Shishesaz, S. M. Kassiriha and M. Nematollahi, Characterization of structure and corrosion resistivity of polyurethane/organoclay nanocomposite coatings prepared through an ultrasonication assisted process, Prog. Org. Coat., 2010, 68, 180–188,  DOI:10.1016/j.porgcoat.2010.02.006.
32. W. Lin, C. Wang, B. Long and Y. Huang, Preparation of acrylic anodic electrophoretic resin/clay nanocomposite films by water-based electrodeposition, Compos. Sci. Technol., 2008, 68, 880–887,  DOI:10.1016/j.compscitech.2007.08.020.
33. C. Song and G. Villemure, Preparation of clay-modified electrodes by electrophoretic deposition of clay films, J. Electroanal. Chem., 1999, 462, 143–149,  DOI:10.1016/s0022-0728(98)00397-0.
34. C.-A. Wang, B. Long, W. Lin, Y. Huang and J. Sun, Poly(amic acid)–clay nacre like composites prepared by electrophoretic deposition, J. Mater. Res., 2011, 23, 1706–1712,  DOI:10.1557/jmr.2008.0209.
35. I. Deen, X. Pang and I. Zhitomirsky, Electrophoretic deposition of composite chitosan–halloysite nanotube–hydroxyapatite films, Colloids Surf., A, 2012, 410, 38–44,  DOI:10.1016/j.colsurfa.2012.06.011.
36. M. Jalilian, H. Yeganeh and M. N. Haghighi, Preparation and characterization of polyurethane electrical insulating coatings derived from novel soybean oil-based polyol, Polym. Adv. Technol., 2010, 21, 118–127,  DOI:10.1002/pat.1406.
37. C. Wan, X. Bao, F. Zhao, B. Kandasubramanian and M. P. Duggan, Morphology and properties of silane-modified montmorillonite clays and clay/PBT composites, J. Appl. Polym. Sci., 2008, 110, 550–557,  DOI:10.1002/app.28672.
38. H. Bakhshi, H. Yeganeh, S. Mehdipour-Ataei, A. Solouk and S. Irani, Polyurethane Coatings Derived from 1,2,3-Triazole-Functionalized Soybean Oil-Based Polyols: Studying their Physical, Mechanical, Thermal, and Biological Properties, Macromolecules, 2013, 46, 7777–7788,  DOI:10.1021/ma401554c.
39. G. Sankar and A. S. Nasar, Cure-Reaction Kinetics of Amine-Blocked Polyisocyanates with Alcohol Using Hot-Stage Fourier Transform Infrared Spectroscopy, J. Appl. Polym. Sci., 2008, 109, 1168–1176,  DOI:10.1002/app.
40. Z. Ranjbar and S. Moradian, Characteristics and deposition behavior of epoxy_amine adducts in cathodic electrodeposition as a function of the degree of neutralization, Colloids Surf., 2003, 219, 147–159,  DOI:10.1016/s0927-7757(03)00028-1.
41. C. Yang and Y. Chen, Preparation and Electrodeposition Properties of Dimethylaminoethanol/2-Ethyl hexanol Resins, Appl. Polym. Sci., 1992, 45, 1055–1063.
42. T. Fqjibayashi, T. Kasukawa and K. Morimoto, A novel binder suitable for cationic electrodeposition coatings based on a non-isocyanate chemistry, Prog. Org. Coat., 1993, 22, 231–244.
43. S. Rastegar and Z. Ranjbar, Cathodically electro-deposited carbon-black-epoxy composite films as primer in two-coat electro-deposition, Prog. Org. Coat., 2006, 56, 234–239,  DOI:10.1016/j.porgcoat.2006.05.009.
44. M. Alam, E. Sharmin, S. M. Ashraf and S. Ahmad, Newly developed urethane modified polyetherimide-based anticorrosive coatings from a sustainable resource, Prog. Org. Coat., 2004, 50, 224–230,  DOI:10.1016/j.porgcoat.2004.02.007.
45. K. Hsu-Ching, C. Wen-Ping and M. Chen-Chi, Synthesis and characterization of a clay/waterborne polyurethane nanocomposite, J. Mater. Sci., 2005, 40, 179–185.
46. K. I. Suresh and M. G. Harikrishnan, Effect of cardanol diol on the synthesis, characterization, and film properties of aqueous polyurethane dispersions, J. Coat. Technol. Res., 2014, 11, 619–629,  DOI:10.1007/s11998-014-9571-1.
47. D. S. Kim, J. Kim and W. B. Woo, Reaction Kinetics and Characteristics of Polyurethane/Clay Nanocomposites, J. Appl. Polym. Sci., 2005, 96, 1641–1647,  DOI:10.1002/app.21616.
48. H. T. Jeon, M. K. Jang, B. K. Ki and K. Kim, Synthesis and characterization of waterborne polyurethane-silica hybrids using sol–gel process, Colloids Surf., A, 2007, 302, 557–567.
49. S. Fang, L. Zhou, L. Gao, D. Liu and H. Zhang, Preparation and characterization of nanocomposite material based on polyurethane acrylate macromonomer, Polym. Compos., 2009, 30, 731–736,  DOI:10.1002/pc.20605.
50. J. Xiong, Y. Liu, X. Yang and X. Wang, Thermal and mechanical properties of polyurethane/montmorillonite nanocomposites based on a novel reactive modifier, Polym. Degrad. Stab., 2004, 86, 549–555,  DOI:10.1016/j.polymdegradstab.2004.07.001.
51. J. Xiong, Z. Zheng, H. Jiang, S. Ye and X. Wang, Reinforcement of polyurethane composites with an organically modified montmorillonite, Composites, Part A, 2007, 38, 132–137,  DOI:10.1016/j.compositesa.2006.01.014.
52. A. Lasia, Electrochemical Impedance Spectroscopy and its Applications, Springer, 2014,  DOI:10.1007/978-1-4614-8933-7.
53. M. F. De Riccardis, V. Martina, D. Carbone, M. Re, E. Pesce and R. Terzi, et al., Incorporation of montmorillonite in epoxy to obtain a protective layer prepared by electrophoretic deposition, Mater. Chem. Phys., 2011, 125, 271–276,  DOI:10.1016/j.matchemphys.2010.09.027.
54. E. Cano, D. Lafuente and D. M. Bastidas, Use of EIS for the evaluation of the protective properties of coatings for metallic cultural heritage: a review, J. Solid State Electrochem., 2010, 14, 381–391,  DOI:10.1007/s10008-009-0902-6.
55. M. Mahdavian and M. M. Attar, Another approach in analysis of paint coatings with EIS measurement: Phase angle at high frequencies, Corros. Sci., 2006, 48, 4152–4157,  DOI:10.1016/j.corsci.2006.03.012.
56. E. Potvin, L. Brossard and G. Larochelle, Corrosion protective performances of commercial low-VOC epoxy/urethane coatings on hot-rolled 1010 mild steel, Prog. Org. Coat., 1997, 31, 363–373.
57. A. Amirudin and D. Thierry, Application of electrochemical impedance spectroscopy to study the degradation of polymer-coated metals, Prog. Org. Coat., 1995, 26, 1–28.
58. R. M. Souto and Y. Gonza, Electrochemical and structural properties of a polyurethane coating on steel substrates for corrosion protection, Corros. Sci., 2007, 49, 3514–3526,  DOI:10.1016/j.corsci.2007.03.018.
59. N. Kouloumbi and S. Kyvelidis, Evaluation of the Anticorrosive Behaviour of Organic Coatings by Using a Variant of Electrochemical Impedance Spectroscopy, Mikrochim. Acta, 2001, 136, 175–180.
60. X. Zhang, F. Wang and Y. Du, Protective performance of epoxy resin modified with coal tar coating studied by electrochemical impedance spectroscopy, Prog. Org. Coat., 2005, 53, 302–305,  DOI:10.1016/j.porgcoat.2005.03.010.
61. N. Kouloumbi, G. M. Tsangaris, A. Skordos and S. Kyvelidis, In Organic Evaluation of the behaviour of particulate polymeric coatings in a corrosive environment. Influence of the concentration of metal particles, Prog. Org. Coat., 1996, 28, 117–124.
62. Y. Zhu, D. Sun, H. Zheng, M. Wei and L. Zhang, Improvement in properties of epoxy-based electrophoretic coating by silica/polyurethane nanocomposites, J. Mater. Sci., 2007, 42, 545–550,  DOI:10.1007/s10853-006-1066-8.

---