High corrosion resistant – redox active – moisture curable – conducting polyurethanes

Progress in Organic Coatings 94 (2016) 79–89

Contents lists available at ScienceDirect

Progress in Organic Coatings journal homepage: www.elsevier.com/locate/porgcoat

High corrosion resistant – redox active – moisture curable – conducting polyurethanes Ravi Arukula a , Ramanuj Narayan a , B. Sreedhar b , Chepuri R.K. Rao a,∗ a b

Polymers and Functional Materials Division, CSIR-Indian Institute of Chemical Technology, Hyderabad 500007, India Inorganic and Physical Chemistry Division, CSIR-Indian Institute of Chemical Technology, Hyderabad 500007, India

a r t i c l e

i n f o

Article history: Received 29 May 2015 Received in revised form 31 December 2015 Accepted 26 January 2016 Keywords: Polyurethane Conductivity Tetraaniline Electrochemical Corrosion

a b s t r a c t New series of electrochemically active conducting polyurethanes (CPU) have been prepared using tetraaniline as conducting unit with IPDI-PTMG as urethane prepolymer segment. These films are robust and formed as a consequence of moisture curing of excess isocyanate. The films are novel in the sense that the conductivity is found to be in the higher side of the order of 10−3 S/cm and the tetraaniline segments are electrochemically oxidizable and/or reducible. The formed films have been subjected for thermal and mechanical tests such as TGA, DSC, DMTA and tensile tests to explore into the understanding of structure–property relations. The nano/microstructure of the polymers has been investigated by TEM analysis. These CPUs are found to be excellent anti-corrosion barrier coatings for mild steel. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Polyurethanes (PU) are unique commercial polymers of this century as these are useful for different applications as adhesives, elastomers, coatings for textiles/paper, foot wear, furniture/foams, packaging material, and for automotive finishes [1–3]. The biocompatible products such as nasogastric catheters, peritoneal dialysis and infusion pumps to implanted pacemaker parts are also derived from polyurethane polymers [2]. PU elastomers are multiblock copolymers comprising of alternate “soft” polyether or polyester segments and “hard” polyurethane segments. The thermodynamic incompatibility of these segments, combined with crystallization of either or both segments, drives their microphase separation into hard and soft phases that are, respectively, below and above their glass transition temperatures. This microphase separation is responsible for the excellent elastomeric properties of polyurethanes [4]. There have been attempts found in literature for converting these insulating polymers into electrically-conducting advanced materials for application such as sensors, corrosion resistance paints, electrostatic dissipaters, EMI shielders, shape memory materials and others [5–7]. Binary composite systems, comprising

∗ Corresponding author. E-mail address: [email protected] (C.R.K. Rao). http://dx.doi.org/10.1016/j.porgcoat.2016.01.022 0300-9440/© 2016 Elsevier B.V. All rights reserved.

of conductive fillers such as carbon black (CB), CNTs, metal powder or conducting polymer in the polymer matrix, resulted in materials that are tough, flexible, and electrically conductive [5–8]. These unique materials, as mentioned earlier, are ideally suited for antistatic layers, electromagnetic interference shielding (EMI), chemical vapour sensors and thermal resistors [9,10]. In contrast to polyaniline, the interesting property of the oligoanilines is their ability to undergo easy processability. Oligoanilines exhibit unique molecular weight and monodipersity and also easily functionalizable. Because of these positive traits, there is a surge in synthesis of oligoaniline-based novel materials which exhibited interesting properties like pH dependent self assembly, tuneable electrochromic property and different nanoscaled morphologies [11–19]. Recently there are attempts to incorporate ‘oligoanilines’ into the back-bone of different types of processable polymers as one end attached hanging pendent group or both-ends inserted redox segment [20–23]. Electrochemical [20] and conductivity studies [21] on poly(methacrylamide) containing four and five aniline units have been reported in the literature. Apart from being electrochemical active, the polymer containing pentaniline as pendant group is electronically conducting to the tune of 10−4 S/cm after doping with mineral acid. Wang et al. [22] have reported polyimides from diamino capped TriAni which are electroactive with good thermal stability. Similar electroactive polyimides have been synthesized and

80

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

studied for their electrochemical activity [23]. The polymer on GC electrode showed three oxidations and three corresponding reductions. The conductivity of the pressed polymer is 1.04 × 10−4 S/cm. Earlier, in our previous communication [24] we incorporated/terminated polyurethane back-bone with tetraaniline or trianiline for a different series of polyurethane-prepolymers. However, the film formation was very slow and the corrosion protection was moderate. The electrochemical properties here not uniform. In the present study, we incorporated tetraaniline (TAni), one of the important oligoanilines, as terminal group in polyurethane by reacting with NCO-terminated pre-polyurethane with some fraction of NCO leaving for moisture curing. These polyurethanes formed excellent films, in contrast to our earlier report [24]. We attempted to throw light on the mechanical, thermal, electrical, electrochemical and anti-corrosion properties of these novel conducting polymeric materials. The results are presented and discussed. 2. Experimental 2.1. Materials Isophorone diisocyanate (IPDI), poly(tetramethylene ether glycol) (PTMEG; Mn = 1000), trimethylol propane (TMP), dibutyltin dilaurate (95%), p-toluene sulphonic acid (PTSA), N-phenyl1,4-phenylenediamine were procured from Aldrich Chemicals, USA. Ferric chloride, hydrochloric acid, acetone, n-butyl amine, toluene and sodium chloride chemicals were obtained from SD FINE Reagents Co., India. Methylethylketone (MEK), hydrochloric acid (36%), ammonium hydroxide, 1-methyl-2-pyrrolidone (NMP) chemicals were purchased from SDFINE Co., India. Tetraaniline (TAni) was prepared [25] according to a literature procedure described by Zang et al.

2.2. Synthesis of conducting polyurethanes (CPUs) The preparations of TAni containing conducting polyurethanes (CPU-10%, CPU-12.5% and CPU-15%) were carried out in two step reaction sequence. First a three-necked, moisture-free round bottomed flask was flushed with dry nitrogen and charged with calculated amount of IPDI (0.027 eq.), PTMEG-1000 (0.008 eq.), TMP (0.0133 eq.) and a drop of dibutyltin dilaurate was added as catalyst. The flask was heated at 70–75 ◦ C for nearly 2–3 h with vigorous overhead stirring. During this time NCO values were checked at constant time intervals by standard n-butyl amine titration method. When appropriate NCO% value was reached, the pre-polymer was reacted with various amounts of TAni (10 wt%, 12.5 wt% or 15 wt% based on PTMG weight or 0.0011 eq., 0.00137 eq. or 0.00165 eq.) after dissolving in small amount of MEK and 1-methyl-2-pyrrolidone (NMP) solvent (10 ml). The reaction mixture was heated with stirring for period of 4 h after which time heating was stopped and stirring continued for further period of 12 h. The tetraaniline containing polyurethane was formed. Pure polyurethane was also synthesized as describe above with 0.0257 eq. of IPDI, 0.008 eq. and 0.0133 eq. of TMP without TAni. The obtained CPUs were casted in Teflon dishes for free standing films. For doped films, 10 g of polyurethane solution from above reaction mixture was stirred with 0.9 g of PTSA for 2 h and casted in Teflon Petri dish for free standing doped films. 2.3. Characterization methods FT-IR spectra were recorded on a Perkin Elmer Spectrum-100 spectrometer using KBr pellet technique. UV–vis spectral measurements were carried out on a Spectro UV-Vis Double Beam PC-8 Scanning Auto cell, LABOMED INC. Thermo gravimetric analyses of polymer samples were carried out using TA Instruments (TGA Q500 V20.8, USA) from ambient to 600 ◦ C under nitrogen

Scheme 1. Scheme showing general structure of CPUs and steps involved in its synthesis.

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

81

Table 1 Formulations used for the synthesis of conducting polyurethanes. S. no

Polymer

IPDI (eq.)

PTMG (eq.)

TMP (eq.)

TANi (eq.)

Moisture cured [NCO] (eq.)

1 2 3 4

Pure-PU CPU-10% CPU-12.5% CPU-15%

0.02568 0.027002 0.027335 0.027671

0.008 0.008 0.008 0.008

0.0133 0.0133 0.0133 0.0133

– 0.0011 0.00137 0.00165

0.004383 0.00460 0.00466 0.00472

Table 2 FT-IR spectral data of the doped and undoped CPUs. Characteristic Absorption Bands

N H str. vibrations C H stretching vibrations Amide I(C O) stretching vibrations Amide II (ıN H +  C N +  C C ) C N Amide III Amide-iv Amide-V

Peak position (cm−1 ) CPU-10-U

CPU-10-D

CPU-12.5-U

CPU-12.5-D

CPU-15-U

CPU-15-D

3315.16 2937.83 1693.92 1597.30 1364.04 1231.25 747.78 682.07

3315.16 2938.59 1696.23 1524.34 1363.69 1236.54 747.35 682.46

3344.80 2932.47 1676.65 1528.49 1362.63 1232.01 742.34 667.62

3311.64 2928.83 1686.35 1510.92 1363.17 1234.43 747.88 681.09

3322.32 2936.29 1693.78 1527.07 1364.27 1237.08 746.30 672.02

3315.16 2937.83 1693.92 1597.30 1364.04 1231.25 747.78 682.07

atmosphere at a heating rate of 10 ◦ C per minute. Differential scanning calorimetry analyses of polymer samples were carried out using TA Instruments (DSC Q100 EXFO Series 2000 USA). DMTA experiments were conducted on TA instruments DMA-Q800 instrument (f = 1 Hz, heat rate = 3 ◦ C/min, tensile mode) from −80 ◦ C to 150 ◦ C. Cyclic voltammetry (CV) was performed using AUTOLAB302N-potentiostat-galvanostat equipped with FRA32M module in a three-electrode electrochemical cell using S.C.E as the reference electrode and platinum wire as the counter electrode. The working electrode was prepared by casting NMP/MIBK solution of the CPU polymer onto the surface of the platinum mesh electrode (1 cm × 1 cm). The CV experiments were carried out in 1.0 M sulphuric acid aqueous solution at a scan rate of 100 mV s−1 . Tafel polarization studies were done in a specially fabricated cell, containing NaCl (3.5 wt%) supporting electrolyte. Tafel plots were obtained by scanning the potential from 200 mV below to 200 mV above Ecorr at a scan rate of 5 mV min−1 . Corrosion current density (Icorr ) and other parameters were obtained directly from the software. Conductivity of the cast films were measured by four probe method using 6220 constant current source and 2182A voltmeter (Keithley, Cleveland, Ohio, USA). Morphological studies for films were performed using Hitachi 3000N, scanning electron microscope operating at 10 kV. The sample was mounted on a double-sided adhesive carbon disk and sputter-coated with a thin layer of gold to prevent sample from possible charging. AFM profiles were pictured on JEOL’s JSPM-5200TM, microscope. The images were recorded in the noncontact mode. All recordings were made in air under ambient conditions to produce 2D and 3D images. TEM experiments were conducted on PHILIPS TECHNAI FE12 instrument at 120 kV after casting a drop of well dispersed dilute samples in water on copper grids. The surface of the samples was analyzed using a KRATOS AXIS 165 X-ray photoelectron spectrometer (UK) at room temperature. The X-ray gun was operated at 15 kV voltage and 20 mA current. Survey and high-resolution spectra were collected using 80 eV and 40 eV pass energy, respectively.

3. Results and discussion The synthesis of the conducting polyurethanes (CPU) is depicted in Scheme 1 and the compositions are shown in Table 1. The synthesis consists of two step process: (i) reaction of IPDI with PTMG-1000 precursors culminating in NCO-terminated prepolymer; (ii) chain termination of the pre-polymer using TAni oligomer in emeraldine base oxidation state. Three conducting polyurethanes, abbreviated

as, CPU-10, CPU-12.5 and CPU-15 were synthesized in this study which contained 10 wt%, 12.5 wt% and 15 wt% TAni as conducting segments. The generic structure of the CPU is shown in Scheme 1. The polymers are obtained as blue coloured viscous liquids with reasonable shelf life (about 80–90 days) after which time they tend to become gel. The polymer solutions were doped with stoichiometric (1:2 mole) p-toluene sulphonic acid before casting into 0.5 mm thick green films. These films (doped and undoped) have been characterized for their spectral, thermal, mechanical, conductive and electrochemical properties using conventional instrumental methods. 3.1. FT-IR spectral characteristics The wet polymers in solution state as well as their films have been tested for their IR-absorption behaviour. The FT-IR spectrum of the pre-polymer formed between IPDI and PTMG-1000 exhibited prominent peak for free-NCO group with an intense absorption at 2260 cm−1 . The intensity of this band decreased after reaction with TAni. The most prominent peaks which do not interfere with other bands from the PU segments only appeared in the hybrid PU-TAni films. The main and distinguishable peaks at 1597 cm−1 and 1500 cm−1 correspond to C C stretching vibrations of quinoid and benzenoid rings of TAni, respectively, are distinguishably seen in Fig. 1. The quinoid band of pure TAni at 1597 cm−1 appeared between 1592 cm−1 and 1597 cm−1 in the CPU films. The peak at 1509 cm−1 due to benzenoid ring of TAni segment is also seen as small peak/hump at 1510 cm−1 . Amide-I ( C O) stretching band originating from polyurethane is seen as broad and intense due to complex contributions due to C O stretching, the C N stretching and the C C N deformation vibrations (Fig. 1, Table 2). The spectra also exhibited Amide II bands between 1534 cm−1 and 1527 cm−1 which is believed to be result of mixed contribution of ıN H +  C N +  C C . Amide III is highly mixed and complicated by coupling with NH deformation modes and is observed between 1231 cm−1 and 1244 cm−1 in the present conducting polyurethane polymers [26–28]. Thus the FT-IR analysis showed that the conducting segment is formed in the hybrid polymers CPU-10, CPU-12.5 and CPU-15. 3.2. UV–vis absorption study UV–vis electronic spectroscopy is a valuable tool for conducting polyaniline (or oligoanilines) and their composites for

82

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

(1:2) doping. There are many NH centres available in polyurethane segments, which tend to abstract the protons from the stoichiometrically added acid, resulting in insufficient doping of quinoid segments of Tani in CPU. Similar observation is found in our earlier communication [24] as well as in literature [29,30]. 3.3. Electrical conductivity measurements

Fig. 1. FT-IR spectral data of conducting polyurethanes.

Table 3 UV–vis data of the polymers in DMF solvent. Polymer

UV–vis (max , nm)a

Conductivity (S/cm) (measured as film)

Pure-PU CPU-10-U CPU-10-D CPU-12.5-U CPU-12.5-D CPU-15-U CPU-15-D

No peaks 527.0 421.0, 618.0 531.0 428.0, 620.0 536.0 432.0, 628.0

∼10−10 – 1.7 × 10−3 –1.86 × 10−3 – 4.7 × 10−4 –2.2 × 10−3 – 8.36 × 10−3 –1.1 × 10−2

a The concentration for the measurements is about 0.5 g of polymer sample in 20 g of DMF solvent.

The structural characteristics of most electro-active polymers are ␲ system extending over a large number of repeating monomer units. This arrangement results in materials with directional conductivity strongest along the axis of the chain. If the polymer chains are not ordered or randomly distributed, an amorphous material with anisotropic properties is resulted. When an electric field is applied on this material, a conductive path is formed and charge carriers are transported from one chain to another. This transfer is called intermolecular hopping and is a thermally assisted tunnelling effect [31]. The conduction critically depends on the hopping between conjugated parts of the polymer. These theories are also applicable to describe the charge transport in conducting and semiconducting organic materials such as conjugated polymers [32–34] oligomers and small molecules [35]. In the present conducting systems, the hybrid films are expected to show electronic conductivity due to TAni units when properly doped. Thus the signature properties of conducting polyaniline, i.e., the electrical conductivity and redox behaviour have also been studied for the cast films. The conductivity behaviour of the samples is understood by measuring current and voltage responses (hence resistance) of the film samples obtained after doping by stoichiometric amounts of PTSA. The conductivity of the polymers was determined using Van der Pauw Eq. (1) by four probe method. =

identifying the conducting and insulating states. All the polymers were characterized by electronic absorption spectroscopy in the range 300–900 nm. The spectra are shown in Fig. 2 and the data are collected in Table 3. TAni exhibited two visible absorption bands at 310 nm and 580 nm in NMP solvent [29]. Upon doping with HCl solution, two new bands emerged at 430 nm and 800 nm. These are attributed to the delocalized polaron bands [29]. The pre-polymer did not show any visible bands in the range studied. The undoped polymers CPU-10, CPU-12.5 and CPU-15 exhibited a clear band at 527 nm, 531 nm and 536 nm, respectively. After doping with PTSA, two bands appeared in the absorption spectra at 421 nm, 618 nm; 428 nm, 620 nm and 432 nm, 628 nm, respectively. Thus the shift from ∼530 nm to 800 nm, as in the case of TAni is not observed in the present hybrid polymers. In the case of TAni, the shift from ∼530 nm to 800 nm is clearly observed due to sufficient doping in HCl solution. But this shift to 800 nm is not observed in the present hybrid polymers due to insufficient doping arising due to stoichiometric

ln 2 i × v d

(1)

Where  is the surface electronic conductivity, d is the thickness of the film (which should be less than distance between two probes), and i and v are the current and voltages applied and measured. The thicknesses of the films cast are in the range 0.33–0.48 mm. The conductivity of the films increased with the increase of TAni fraction in the polymer which is understood as due to increased concentration of polarons formed. Thus the range of conductivity of these films is found to be in the range 1.7 × 10−3 –1.1 × 10−2 S/cm (Fig. 3 and Table 3). The range is suitable for ant-static, electrostatic dissipation and electromagnetic interference applications. The hopping mechanism for conduction has been schematically shown in our previous article [24]. The generated charge carriers move along the limited conjugated chain and hop to the nearest TAni segments causing conduction in the film. The films showed small scale variations in the surface conductivity values which can

Fig. 2. UV–vis spectra of the conducting polyurethanes in DMF solvent.

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

83

Table 4 XPS data of the conducting polyurethanes.

Fig. 3. Conductivity variation profile of films obtained from CPUs.

Polymer

C1s (B.E, eV)

Pure-PU

280.78 281.91 282.76 284.49

527.88 529.54 – –

O1s (B.E, eV)

N1s (B.E, eV) 396.27 397.02 399.34

CPU-10%-U

284.60 285.60 286.43 288.01

531.69 533.02 – –

397.43 399.43 400.86 401.86

CPU-10%-D

284.60 285.75 286.94 288.30

531.97 533.59 – –

397.73 399.15 400.41 401.86

be explained on the fact that the concentration of the oligoaniline units are not uniformly distributed when the film formed. 3.4. Microscopy 3.4.1. Structure and morphology of the films by SEM and AFM The surface morphology of the films has been explored by scanning electron microscopy (SEM; Fig. 4). The unmodified, moisture cured pure PU film exhibited smooth surface with line type pattern on the surface (Fig. 4(g)). The 10%-TAni modified PU film is also smooth but without any line type pattern. Few particles of about 1–2 micron size are present on both the films (Fig. 4(a)–(c)). However, the smoothness of the undoped film decreased (roughness increased) considerably on doping with PTSA (Fig. 4(d)–(f)). This is understood as the result of ion of PTSA anion into the polymer matrix for charge neutrality. This is also very clearly observed in AFM study. Fig. 5(a) and (b) compares the AFM pictures of undoped and doped CPU-10 films. The mapping on the undoped film showed medium to larger islands spread over 2–3 ␮m range, which are basically flat. In the doped film, sphere/oval type particles are seen due to ingression of PTSA. The surface roughness of the undoped film is about 300 nm where as it is about 447 nm after doping. 3.4.2. TEM study The CPU-10 and CPU-12.5 were characterized by tunnelling electron microscope (TEM) in both doped and undoped form. For this purpose the undoped polymer solutions were dispersed in water and sampled. The TEM profiles are presented in Fig. 6. The undoped polymer formed platelet type particles with different diameters such as 38 nm, 81 nm, 113 nm, 124 nm and maximum of 135 nm on the TEM grid (Fig. 6(a)–(d)). These platelets are entangled to each other to form platelet-islands. The dispersibility of these polymers into platelets is considerably lowered in presence of doping acid, HCl. The round platelets of undoped polymer changed into irregularly shaped, entangled and diffused structures as shown in Fig. 6(e) and (f). This is thought to be due to the ingression of ions into the polymer matrix while doping process. 3.5. XPS study X-ray photoelectron spectroscopy (XPS) has been very valuable technique for characterization of the polyurethane surfaces [36,37]. XPS studies showed that enrichment in soft segments on the airside surface of polyurethanes [38–41]. The enrichment of the softsegment resulted in enhancement of microphase separation in the bulk [38–40]. Almost all the XPS investigations on polyurethanes available in literature are focussed on C1s and O1s spectra with

Scheme 2. Moisture curing reactions of

NCO group.

scant attention on N1s core. In this study, we tried to focus on all the three elements. The C1s spectra (Fig. 7, Table 4) showed four resolvable peaks which are assignable to functional groups of C C/C H, C O, C O and C N, respectively, on the surface of the CPU [37]. The O1s spectra can be deconvoluted into to two-major contributors, oxygen’s of carbonyl and ester type at 531.7 eV and 533 eV, respectively. There are no reports available on the XPS data of TAni. However, literature reports of on emeraldine base (EB) of polyaniline suggested that the N1s core level XPS profiles can show two nearequal intensity peaks assignable to quinoid ( N ) and benzenoid ( NH ) structures, respectively, at 398.1 eV and 399.3 eV [42]. In the doped state with HCl, the N1s spectrum showed three peaks, one peak more than pure EB, at about 401.0 eV which is assignable to charged species N+ . It was observed that the intensity of imine peak reduced while the intensity of peak due to the species N+ increased on increasing doping levels [42–44]. In general the N1s core spectrum of polyurethanes exhibits one peak due to single type of NH species from urethane linkage at 398.9 eV [45]. As explained in Scheme 2, the presently studied moisture cured polyurethanes contain two types of nitrogens: (i) NH2 (amino group) and (ii) NH CO NH (urea group) obtained as result of moisture curing reactions. Apart from the above two groups, the third functionality (iii) NH CO O (urethane group) from pre-polymer will also contribute to the XPS spectrum. These three species resulted in two main N1s peaks. The lower energy peak with higher intensity can be resolved into two peaks at 396.27 eV and 397.02 eV assignable to (i) and (ii), respectively. The other peak at higher energy 399.35 eV is due to the species (iii). The N1s spectrum for CPU-10 doped is shown in Fig. 7(c). The three peaks (i)–(iii) appeared in pure polyurethane peak appeared with shift in binding energy towards higher side at 397.71 eV, 399.44 eV and 400.39 eV. In addition to this there is fourth peak appeared at 401.86 eV which is possibly arising due to N or N+ species from tetraaniline segments. The XPS data are collected in Table 4.

84

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

Fig. 4. Scanning electron micrographs (SEM) of CPU-10-undoped at: (a) 1 ␮m; (b) 2 ␮m; (c) 10 ␮m; CPU-10-doped at: (d) 1 ␮m; (e) 2 ␮m; (f) 10 ␮m; pure polyurethane at: (g) 10 ␮m

Table 5 DMTA and DSC data of the samples. Sample name

Pure-PU CPU-10%-U CPU-12.5%-U CPU-15%-U NC, not clear.

Storage modulus (SM)

Loss modulus (LM)

Tan (ı) ( ◦ C)

SM (MPa)

Temperature ( ◦ C)

LM (MPa)

Temperature ( ◦ C)

Tg1 ( ◦ C)

2571 – 2154 – 2725 – 2532 –

−86.90 – −82.88 – −85.48 – −85.46 –

141.1 55.32 95.1 68.96 116.6 93.0 123.0 86.42

−48.38 25.83 −39.16 14.96 −34.67 11.88 −37.98 15.67

−41.52 – −40.34 – NC – −27.11 –

DSC Tg2 ( ◦ C) 62.69 – 74.27 – 69 – 64.33 –

Tg1 ( ◦ C)

Tg2 ( ◦ C)

NC – −34.6 – −15.4 – −2.0 –

27.7 – 44.2 – 47.9 – 42.4 –

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

85

Fig. 5. AFM profiles of the CPU-10 film. (a) Undoped; (b) doped film.

Fig. 6. TEM pictures of the conducting polyurethanes. (a) CPU-10-undoped@1 ␮m; (b) CPU-12.5-undoped@200 nm; (c) CPU-10-undoped@1 ␮m; (d) CPU-12.5undoped@1 ␮m; (e) CPU-10-HCl doped@2 ␮m; (f) CPU-10-HCl doped@1 ␮m.

86

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

Fig. 7. XPS profiles of CPU-10-doped.

Table 6 TGA data of the conducting polyurethanes (CPUs). Sample

T1ON

T1MAX

T2ON

TF

Weight percentage at T2ON (%)

PURE-PU CPU-10%-U CPU-12.5%-U CPU-15%-U

283.32 265.12 270.90 279.54

329.64 322.40 323.48 328.22

366.51 363.81 369.52 376.91

428.23 450.02 452.23 453.10

42.77 47.51 46.48 45.24

Table 7 Electrochemical and corrosion data of PU and CPUs. Polymer

PURE-PU CPU-10%-U CPU-12.5%-U CPU-15%-U

CV

Tafel(Corrosion)

Oxidation (V, A) Oxd-1

Oxd-2

Reduction (V, A) Red-1

Red-2

Icorr (A/cm2 )

Ecorr (mV, SCE)

Rp (MOhm/cm2 )

CR (mm/year)

– 0.467, 7.30 × 10−7 0.460, 2.36 × 10−6 0.461, 3.87 × 10−6

– 0.697, 1.41 × 10−6 0.675, 5.19 × 10−6 0.677, 6.14 × 10−6

– 0.412, −1.79 × 10−6 0.382, −4.90 × 10−6 0.390,−5.40 × 10−6

– 0.631, −1.14 × 10−8 0.623, −5.10 × 10−7 0.622, −4.51 × 10−7

6.161 × 10−9 3.920 × 10−8 2.068 × 10−9 1.806 × 10−9

−581.73 −529.10 −203.69 −495.59

24.100 3.700 42.511 16.050

7.15 × 10−5 4.55 × 10−4 2.40 × 10−5 2.09 × 10−5

3.6. Thermal and mechanical characterizations 3.6.1. Differential Scanning calorimetry (DSC) The casted films were investigated by DSC (Table 5) and Fig. 8 shows the DSC curves for pure and three other CPU polymers.

The flexible pentamethylene segments acts as soft segment in the polymer chain showed a Tg1 at −34.6 ◦ C and hard isocyanate based hard segment showed a second glass transition (Tg2 ) at 44.2 ◦ C. The first glass transition temperature shifted to higher values as the TAni% increased progressively. This is possibly due to increase

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

87

the second step), TF (final decomposition temperature), percentage weight loss at T2ON are collected in Table 6. The TGA graph clearly showed that the Tdecon for pure polymers is 248 ◦ C and this temperature increased for CPU-10%, CPU-12.5% and CPU-15% indicating increasing thermal stability for the tetraaniline-modified CPUs. The first stage of decomposition accelerates at about 298–301 ◦ C for the polymers and is complete by 390–411 ◦ C. There is about 42–47% material is remaining at the completion of first stage decomposition. The second stage decomposition starts at as low as 363 ◦ C and ends by 453 ◦ C for the samples. Overall the study indicated that the stability of the CPUs increased marginally with incorporation of TAni.

Fig. 8. DSC curves of (a) pure PU; (b) CPU-10; (c) CPU-12.5; (d) CPU-15.

in less flexible TAni concentrations in the polymers, i.e., decreased in the total flexibility of the soft segment. The CPU-12.5% and CPU15% showed a Tg1 −15.4 ◦ C and −2.0 ◦ C, respectively. The Tg2 values did not underwent much influence of TAni concentration upon its increase and showed Tg2 at 44.2 ◦ C, 47.9 ◦ C and 42.4 ◦ C, respectively. The thermal gravimetric analysis of the polymeric films showed two way thermal decomposition of the polymers as suggested by the presence of two peaks in the derivative curves [46]. The TGA profiles are shown in Fig. 9 and the values of T1ON (initial decomposition temperature for the first step of decomposition), T2ON (initial decomposition temperature for the second step of decomposition), T1MAX (temperature of maximum rate of weight loss for the first step), T2MAX (temperature of maximum rate of weight loss for

3.6.2. Dynamic mechanical testing analysis (DMTA) The dynamic mechanical and thermal analyser (DMTA) is an excellent tool to study the relaxation behaviour, change in loss or storage modulus and glass transition temperature (Tg ) of the polymeric materials. In case of urethane polymers the structure, concentration and organization of the hard/soft segments and their interaction have a dominant influence on the physical and mechanical properties. DMTA measures the response or a deformation of a material to periodic varying forces. This test has advantage over other mechanical tests that these tests can be done over a wide range of temperature in a short time and from the results the overall performance of materials can be predicted. The temperature dependent storage modulus (E ), loss modulus and tan ı responses of CPU are presented in Fig. 10 and the data are presented in Table 5. The moisture cured pure polymer film showed excellent storage modulus of 2571 MPa without TAni segments. This decreased considerably (2154 MPa, 16.3%) upon 10% modification in the backbone. The storage modulus recovers positively (2725 MPa, 106%) on further addition of 12.5% of TAni and finally slightly decreased (2532 MPa, 1.5%) on incorporation of 15% TAni. Overall it can be

Fig. 9. Thermal decomposition profiles of (a) pure PU; (b) CPU-10-undoped; (c) CPU-12.5-undoped; (d) CPU-15-undoped polymers

88

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

Fig. 10. DMTA profiles (a) storage modulus; (b) loss modulus; and (c) tan ␦ with temperatures of PU and CPU films.

stated that the storage modulus increased marginally on addition of 12.5% TAni. On the other hand, the loss modulus also suffered a continuous decrease on the modification of the back-bone from 10% to 15%. Tan ı curves showed two characteristic glass transition tempera Tg1 and Tg2 which are due to soft and hard segments. However these two temperatures varied drastically compared to the DSCmeasured values. This variation is ascribed to their basic difference in measurement techniques.

3.7. Electrochemical and anti-corrosion properties The back bone of the polymers contains different amounts (10%, 12.5% and 15%) of tetraaniline which is responsible for its electrochemical activity. The CPUs casted as thin films (about 7 mg/1 cm2 ) on platinum mesh showed two-step – one electron electrochemical oxidation and reduction in 1 M HCl solution. The first oxidation (Fig. 11) took place at 0.464 V and the second oxidation at 0.670 V while reverse of the scan resulted in two successive reductions at 0.632 V and 0.423 V. The currents of oxidation and reduction process increased with increase in TAni% in the polyurethane polymers. The (Eox − Ered ) values are close to 60 mV for both the redox peaks of all three samples indicating that the electron transfer is reversible. These oxidations are corresponding to the transitions from LEB to EB oxidation state and EB to pernigraniline oxidation state, respectively [47]. The electrochemical data are collected in Table 7. These conducting polyurethanes were also explored for their anti-corrosion property on mild steel. The CPUs were coated on uniformly about 90–100 ␮m on 1 inch × 1 inch mild steel panels

Fig. 11. Cyclic voltammograms of the CPUs (a) CPU-10; (b) CPU-12.5; (c) CPU-15 at a scan rate 100 mV s−1 .

and were left for several days for moisture curing. The dried, cured panels were tested for Tafel polarization in 3.5% NaCl solution. The polarization curves are shown in Fig. 12 and the corrosion data from the polarization studies are collected in Table 7. From the data, it is understandable that the corrosion protection of the CPU-12.5% and CPU-15% is better than pure PU. The corrosion rate decreased from 7.15 × 10−5 mmpy for pure PU to 2.40 × 10−5 mmpy (CPU12.5%) and 2.09 × 10−5 mmpy (CPU-15%) signifying the importance of these polymers. In the study, CPU-10% system showed higher current than pure PU coating, suggesting that the corrosion rate (CR = 4.55 × 10−4 mmpy) is higher (lower protection) for CPU-10%,

R. Arukula et al. / Progress in Organic Coatings 94 (2016) 79–89

89

References

Fig. 12. Tafel polarization curves of the CPUs coated on mild steel. (a) Bare mild steel panel; (b) pure PU; (c) CPU-10; (d) CPU-12.5; (e) CPU-15.

compared to pure PU (CR = 7.15 × 10−5 mmpy). This is due to insufficient formation of conductive networks with lower loading (10%) of TAni. When the wt% of TAni increased to 12.5% and 15%, the system resulted in formation of electroactive networks to reduce the corrosion rate lower than pure PU coating. 4. Conclusions First time in literature, we synthesized and characterized electrochemically active conducting polyurethanes. These are obtained by modification of PU back-bone with tetraaniline (10%, 12.5% and 15%) and are moisture curable to give very strong films. The structure, thermal and mechanical properties of the films have been explored. The mechanism of conductivity, electron transfer behaviour and film formation is explained. The films showed excellent storage modulus to the tune of 2725 MPa. The films are electroactive akin to TAni, giving scope to many other applications and are also conducting to the tune of 10−2 –10−3 S/cm when doped with p-toluenesulphonic acid. The films work as very good corrosion protecting coating on mild steel. The best protection is exhibited by CPU-15% with corrosion rate suppressed to 2.09 × 10−5 mmpy. The procedure described in this article gives a method to synthesize/fabricate electron conducting, electroactive and moisture curable polyurethanes or films. Acknowledgments Ravi Arukula acknowledges the Senior Research Fellowship (SRF) from CSIR India. All authors also acknowledge the financial support from the CSIR-network project INTELCOAT (CSC-0114). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.porgcoat.2016. 01.022.

[1] G. Oertel, Polyurethane Handbook, Hanser, New York, 1985, pp. 629. [2] M. Szycher, Handbook of Polyurethanes, CRC Press, Massachusetts, 1999, pp. 696. [3] D.K. Chattopadhyay, K.V.S.N. Raju, Prog. Polym. Sci. 32 (2007) 352–418. [4] V. Sachin, S.L. Cooper, Macromolecules 31 (1998) 9181–9192. [5] M. Moniruzzaman, K.I. Winey, Macromolecules 39 (2006) 5194–5205. [6] L. Bokobza, Polymer 48 (2007) 4907–4920. [7] J.C. Grunlan, W.W. Gerberich, L.F. Francis, J Appl. Polym. Sci. 80 (2001) 692–705. [8] T. Gurunathan, C.R.K. Rao, R. Narayan, K.V.S.N. Raju, Prog. Org. Coat. 76 (2013) 639–647. [9] T. Gurunathan, C.R.K. Rao, R. Narayan, K.V.S.N. Raju, J. Mater. Sci. 48 (2013) 67–80. [10] J. Njuguna, K. Pielichowski, J. Mater. Sci. 39 (2004) 4081–4094. [11] S. Wang, E.B. Berda, X.F. Lu, X.F. Li, C. Wang, D. Chao, Macro Rapid Commun. 34 (2013) 1648–1653. [12] Z. Shao, Y. Zai, H. Jianchen, S. Chandrasekaran, M.L. David, Z. Wei, C.F.J. Faul, J. Mater. Chem. 22 (2012) 16230–16234. [13] D. Chao, R. Yang, X. Jia, H. Liu, S. Wang, C. Wang, E.B. Berda, J. Polym. Sci. A: Polym. Chem. 50 (2012) 2330–2336. [14] C.U. Udeh, F. Natalie, C.F.J. Faul, J. Mater. Chem. 21 (2011) 18137–18153. [15] Y. Wang, H.D. Tran, R.B. Kaner, Macro Rapid Commun. 32 (2011) 35–49. [16] Y. Liu, J. Hu, X. Zhuang, P. Zhang, X. Chen, Y. Wei, W. Xianhong, Macromol. Biosci. 11 (2011) 806–813. [17] Y. Wang, H.D. Tran, L. Liao, X.F. Duan, R.B. Kaner, J. Am. Chem. Soc. 132 (2010) 10365–103373. [18] H. Kim, J.W. Park, J. Mater. Chem. 20 (2010) 1186–1191. [19] W. Xiong, H.F. Wang, Y. Han, Macromol. Rapid Commun. 31 (2010) 1886–1891. [20] R. Chen, C.B. Benicewicz, Macromolecules 36 (2003) 6333–6339. [21] S. Liu, Z.T. Kaizheng, Y. Zhang, Y. Zhu, T.X. Xu, Mater. Lett. 59 (2005) 3715–3719. [22] Z.Y. Wang, C. Yang, J.P. Gao, J. Lin, X.S. Meng, Macromolecules 31 (1988) 2702–2704. [23] D. Chao, M. Xiaobo, L. Xiaofeng, C. Lili, M. Hui, W. Zhang, Y. Wei, J. Appl. Polym. Sci. 104 (2007) 1603–1608. [24] A. Ravi, T. Appa Rao, C.R.K. Rao, R. Narayan, B. Sreedhar, J. Appl. Polym. Sci. 131 (2014) 40794. [25] W.J. Zhang, J. Feng, A.G. MacDiarmid, A.J. Epstein, Synth. Metals 84 (1997) 119–120. [26] Y. Mido, Spectrochim. Acta A: Mol. Spectrosc. 28 (1972) 1503–1518. [27] M.M. Coleman, K.H. Lee, D.J. Skrovanek, P.C. Painter, Macromolecules 19 (1986) 2149–2157. [28] C. Wilhelm, J.L. Gardette, Polymer 39 (1998) 5973–5980. [29] Z. Yang, X. Wang, Y. Yang, Y. Liao, Y. Wei, X. Xie, Langmuir 26 (2010) 9386–9392. [30] S.A. Chen, H.T. Lee, Macromolecules 26 (1993) 3254–3261. [31] W.M. Blom, M.C.J.M. Vissenberg, Mater. Sci. Eng. R: Rep. 27 (2000) 53–94. [32] B. Movaghar, Philos. Mag. B 65 (1992) 811–816. [33] M.A. Abkowitz, Philos. Mag. B 65 (1992) 817–829. [34] C. Tanase, E.J. Meijer, P.W.M. Blom, D.M. Leeuw, Phys. Rev. Lett. 91 (2003) 216601–216604. [35] L.B. Schein, Philos. Mag. B 65 (1992) 795–810. [36] M.J. Hearn, B.D. Ratner, D. Briggs, Macromolecules 21 (1988) 2950–2959. [37] A.K. Mishra, D.K. Chattopadhyay, B. Sreedhar, K.V.S.N. Raju, Prog. Org. Coat. 55 (2006) 231–243. [38] S.C. Yoon, B.D. Ratner, Macromolecules 19 (1986) 1068–1079. [39] S.C. Yoon, B.D. Ratner, Macromolecules 21 (1988) 2392–2400. [40] K. Nakamae, T. Nishino, S. Asaoka, S. Sudaryanto, Int. J. Adhes. 16 (1996) 233–239. [41] T.L. Wang, F.J. Huang, Macromol. Rapid Commun. 20 (1999) 497–504. [42] K.L. Tan, B.G.T. Tan, E.T. Kang, K.G. Neoh, Phys. Rev. B 39 (1989) 8070. [43] W.R. Saleneck, I. Lundstrom, T. Hjertberg, C.B. Duke, E. Conwell, A. Paton, A.G. MacDiarmid, N.L.D. Somasiri, W.S. Huang, A.F. Richter, Synth. Met. 18 (1987) 291–296. [44] H.S. Munro, D. Parker, J.G. Eaves, Springer Series in Solid State Sciences, vol. 76, Springer-Verlag, Berlin, 1988, pp. 257. [45] G. Beamson, D. Briggs, Resolution XPS of Organic Polymer, Wiley, New York, 1992. [46] D.K. Chattopadhyay, B. Sreedhar, K.V.S.N. Raju, J. Appl. Polym. Sci. 95 (2005) 1509–1518. [47] L. Chen, Y. Yu, H. Mao, X. Lu, W. Zhang, Y. Wei, Mater. Lett. 59 (2005) 2446–2450.