Hyperbranched polyol decorated carbon nanotube by click chemistry for functional polyurethane urea hybrid composites

Article abstract of DOI:10.1039/c4ra02442g

In the present work we report a facile decoration of multi-walled carbon nanotubes with hyperbranched polyether polyol using copper(i) catalyzed azide-alkyne click reaction in order to create hydroxy terminal groups. This decoration has been designed to improve the dispersibility of CNTs in the polymer matrix. These hydroxy functional decorated CNTs were then dispersed into poly(tetramethylene ether) glycol (PTMG) at different weight percentages to get the hybrid pre-polymers. These pre-polymers were reacted with 1-isocyanato-4-[(4-isocyanatocyclohexyl) methyl] cyclohexane (H ₁₂-MDI) at a NCO : OH ratio of 1.2 : 1 and cured under atmospheric moisture to get the functional polyurethane-urea-CNT hybrid composites. There has been substantial improvement in the thermal stability, mechanical strength, corrosion resistance and antimicrobial activity of the polyurethane hybrid composites with the increase in carbon nanotube loading in pre-polymers. For example, with 2 wt% loading of carbon nanotubes, the tensile strength of the polyurethane hybrid composite improved from 1.25 N mm^-2 to 6.25 N mm^-2; the water contact angle improved from 54° to 108° and also the rate of corrosion reduced from 0.047 mm per year to 0.0019 mm per year. We also observed that these hybrids possess remarkable shape recovery properties. These results demonstrate that the decorated CNTs can be used as high performance additives for improving various properties of polyurethane hybrids in cost effective and eco-friendly ways.

Full text of DOI:10.1039/c4ra02442g

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Hyperbranched polyol decorated carbon nanotube
by click chemistry for functional polyurethane urea
hybrid composites†
Cite this: RSC Adv., 2014, 4, 24420
Sasidhar Kantheti,a Rohit Ranganathan Gaddam,ab Ramanuj Narayana
a
*
and K. V. S. N. Raju
In the present work we report a facile decoration of multi-walled carbon nanotubes with hyperbranched
polyether polyol using copper(I) catalyzed azide–alkyne click reaction in order to create hydroxy terminal
groups. This decoration has been designed to improve the dispersibility of CNTs in the polymer matrix.
These hydroxy functional decorated CNTs were then dispersed into poly(tetramethylene ether) glycol
(PTMG) at different weight percentages to get the hybrid pre-polymers. These pre-polymers were
reacted with 1-isocyanato-4-[(4-isocyanatocyclohexyl) methyl] cyclohexane (H12-MDI) at a NCO : OH
ratio of 1.2 : 1 and cured under atmospheric moisture to get the functional polyurethane–urea–CNT
hybrid composites. There has been substantial improvement in the thermal stability, mechanical
strength, corrosion resistance and antimicrobial activity of the polyurethane hybrid composites with the
increase in carbon nanotube loading in pre-polymers. For example, with 2 wt% loading of carbon
nanotubes, the tensile strength of the polyurethane hybrid composite improved from 1.25 N mmÀ2 to
6.25 N mmÀ2; the water contact angle improved from 54ꢀ to 108ꢀ and also the rate of corrosion
reduced from 0.047 mm per year to 0.0019 mm per year. We also observed that these hybrids possess
remarkable shape recovery properties. These results demonstrate that the decorated CNTs can be used
as high performance additives for improving various properties of polyurethane hybrids in cost effective
and eco-friendly ways.
Received 20th March 2014
Accepted 21st May 2014
DOI: 10.1039/c4ra02442g
www.rsc.org/advances
counterpart since they have adjustable transition temperature,
highly stable shape, easy shaping, cost effective and easy
processibility.10–12 Shape memory polymer materials have wide
1. Introduction
Carbon nanotubes (CNTs) are among the most researched
allotropes of carbon in the past decade because of their high
aspect ratio, unique structure, and outstanding electronic,
thermal, optical and mechanical properties.1–4 Due to their
excellent mechanical properties, a very small amount of incor-
poration of CNTs into a polymer matrix could lead to hybrid
composites having superior physico-chemical, mechanical,
thermal and bacterial resistance properties.5–9 In general,
incorporation of CNTs into polymer matrices not only improves
the overall physico-chemical properties of the composite but
also inculcates additional properties like shape recovery,
hydrophobicity and anti-corrosive behaviors. Shape recovery
polymers have an edge over their shape-memory alloy
range of applications as coating materials, elastomeric mate-
rials, medical applications, actuators, textiles and many more.
Polyurethane based shape memory polymers are amongst the
most researched one, owing to its hard and so segments which
are microphase separated heterogeneous structure. The so
segment of the polymer helps in xation of the shape acting as a
molecular switch whereas the hard segment which has an
inherently high melting point enables effective cross-linking. In
the view point of shape recovery materials, CNTs could ll the
role of being very effective ller for enhancing high shape
recovery and shape recovery force. This property is attributed
again to the CNT's high thermal conductivity coefficient and
excellent mechanical properties.
The improvement of composite properties depends directly
on the dispersibility and interfacial interactions of CNT with
polymer matrix.13 The dispersion of pristine CNTs is very diffi-
cult due to their insolubility in both organic and inorganic
solvents. Therefore, surface modications are made in order to
make CNTs processable and soluble.14 The surface of multi-
walled carbon nanotubes (MWCNTs) and single-walled carbon
nanotubes (SWCNTs) can be functionalized by both
aDivision of Polymers & Functional Materials, CSIR-Indian Institute of Chemical
Technology, Hyderabad-500007, India. E-mail: kvsnraju@iict.res.in; drkvsnraju@
gmail.com
bAmity Institute of Nanotechnology, Amity University, Noida-201301, India
† Electronic supplementary information (ESI) available: Tables related to TGA,
Raman, solubility, gures related to FT-IR, 1H-NMR, 13C-NMR, ESI-MS, XRD,
FE-SEM, solubility test, UTM, contact angle, antimicrobial result. See DOI:
10.1039/c4ra02442g
24420 | RSC Adv., 2014, 4, 24420–24427
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
covalent15–17 and non-covalent18–21 means. A non-covalent func- poly (tetramethylene ether) glycol (PTMG) matrix. These pre-
tionalization involves surface modication of CNTs via p–p polymers were reacted with 1-isocyanato-4-[(4-iso-
staking interactions, van der Waal's interaction, physical cyanatocyclohexyl) methyl] cyclohexane (H12-MDI) at NCO : OH
entanglement, electrostatic adsorption or hydrogen bonding ratio of 1.2 : 1 and cured under atmospheric moisture to get the
without disturbing the extended p-conjugation of the nano- functional polyurethane–urea–CNT hybrid composites. The
tubes.22 In contrast, a covalent functionalization requires resulting polyurethane–urea–CNT hybrids along with pure
formation of strong chemical bond between CNTs and the polyurethane were analyzed by TGA, universal testing machine
moieties to be attached. Oxidation of CNTs is the most simple (UTM), atomic force microscopy (AFM), XRD, FT-IR, FE-SEM,
covalent modication which terminates the CNT surface with a water contact angle, antimicrobial and electrochemical polari-
carboxyl group (–COOH) or hydroxyl groups (–OH). The oxida- zation studies.
tion of CNT can be achieved by either direct exposure of CNTs to
xed air or oxygen, treatment with nitric acid, mixture of
aqueous hydrogen peroxide and sulphuric acid, acid mixtures
etc.23 It is very pertinent to have –COOH group, because of its
2. Experimental section
2.1. Materials
versatility in providing many different chemical reactions to
occur on CNT surface.24 Further, the –COOH group could be
modied with various molecules to obtain the necessary surface
termination. The mechanical, thermal and optical properties
depend on the p conjugations of CNT.
Pristine multi walled CNTs (purity 95%), 15–20 nm in diameter
(PlasmaChem GmbH) were used without further purication.
Trimethylolpropane (TMP), propargyl alcohol, boron triouride
diethyl etherate, copper(I) iodide were purchased from Aldrich
Chemicals (Milwaukee, WI, U.S.). Epichlorohydrin, acetonitrile,
sodium azide, chloroform, anhydrous triethylamine, tetrahy-
drofuran (THF), dimethyl formamide, tetra butyl ammonium
bromide (TBAB), diisopropyl ethylamine (DIPEA), ammonia
solution (NH3$H2O) were purchased from Finar Reagents,
Mumbai, India.
The covalent modication of CNTs with hyperbranched
polymers (HBPs) is one of the promising ways to impart supe-
rior properties to CNTs. These branched polymers are macro-
molecules with three-dimensional architecture, which possess
excellent solubility, low melt viscosity, broader molar mass
distribution and extremely high density of functional groups at
the surface.25 A major advantage of HBPs is that, they possess
large number of terminal functional groups which could be
further exploited by coupling them with other materials like
uorophores, dyes and other electroactive groups.26 Anchoring
HBPs on CNTs could be achieved by a variety of well established
reactions. Click chemistry has been used to attach liner poly-
mers to CNTs, however, HBP attachment is still unexplored.
Among all click reactions, copper(I) catalyzed azide–alkyne 1,3-
dipolar cycloaddition reaction is most popular in polymer and
material science because it provides high quantitative yield,
high tolerance of functional groups,27 simple reaction condi-
tions and insensitivity to solvents and reaction medium,28 high
chemo selectivity with no side products, chemical inertness of
1,2,3-triazole ring and ease of the product isolation. Thus,
employing click chemistry is a feasible approach for surface
functionalization of CNTs with polymer.29
Hence, in the present study we report a novel and easy
approach for the modication of CNT with hyperbranched
polymer using click chemistry for the preparation of functional
polyurethane–urea–CNT hybrid composites. Initially the
carboxyl terminated CNTs (CNT–COOH) were converted to
alkyne terminated CNT (CNT–alkyne) by esterication with
propargyl alcohol. The obtained CNT–alkyne was treated with
an azide terminated hyperbranched polyol (HBP–N3) by click
reaction in order to obtain hydroxyl terminated and hyper-
2.2. Characterizations
Fourier transform IR spectra (FTIR) of synthesized samples
recorded by Thermo Nicolet Nexus 670 spectrometer. The 1H
and 13C NMR of the synthesized samples were done inVARIENE-
200 and BRUKER-300 MHz spectroscopy by taking tetra methyl
silane (TMS) as standard at room temperature and dissolving in
DMSO-d6 solvent. Mass spectra (ESI-MS) of synthesized hyper-
branched polyols were recorded on a LC-MSD-Trap-SL mass
spectrometer. XRD patterns for the all the samples were
obtained using a Siemens D-5000 X-ray diffractometer with Cu
Ka radiation of wavelength 1.54. The thermogravimetric anal-
ysis was conducted on TGA Q500 Universal TA instrument (U.K)
at temperature ramp rate of 10 ꢀC minÀ1 from 25 to 600 ꢀC with
a continuous N2 ow at the rate of 30 ml minÀ1. Raman spectra
was recorded using Horiba JobinYvon Raman spectrometer
with a laser excitation wavelength of 632.81 nm. The changes in
morphology of MWCNTs due to surface modication were
observed under FESEM using S4300 SEIN HITACHI Japan at
10 kV. The samples were coated with a thin gold layer of ꢁ5 nm
of thickness by sputtering process to make them conducting for
SEM analysis.
2.3. Synthesis of chlorinated hyperbranched polyether
polyol (HBP–Cl)
branched polyol decorated CNT (CNT–HBP). The various steps The chlorinated hyperbranched polyol was prepared according
involved in surface modication were conrmed by thermog- to a previous report.30 Trimethylolpropane (2 g, 14.9 mmol,
ravimetric analysis (TGA), eld emission scanning electron 1 eq.) was taken into a single neck round bottom ask equipped
microscopy (FE-SEM), transmission electron microscopy (TEM), with nitrogen inlet and a magnetic stirrer. Initially TMP was
X-ray diffraction (XRD), Fourier transform infrared (FT-IR) and heated at 56 ꢀC up to melt. Aer complete melting the heat was
Raman spectroscopic techniques. The obtained CNT–HBP was removed and 0.01 ml of BF3 etherate was added as a catalyst. To
loaded in different weight percentages such as 0.5, 1 and 2 into this reaction mixture epichlorohydrin (4.13 g, 44.7 mmol, 3 eq.)
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 24420–24427 | 24421

View Article Online
RSC Advances
Paper
was added drop wise for a period of 10 minutes. Aer complete resulting CNT–COOH (0.2 g), excess thionyl chloride (4 ml) was
ꢀ
addition the reaction mixture was stirred at 70 C for 12 h.
added in a round bottom ask and then stirred for 24 hours at
FT-IR (KBr, cmÀ1): 3432.78 (O–H str), 2956.04 (asym –CH3 70 C respectively. Aer this, excess SOCl2 was removed under
ꢀ
str), 2917.58 (asym –CH2 str), 2879.12 (sym –CH3 str), 1451.57 reduced pressure. Then the ask was cooled by an ice bath. To
(C–O str), 1100–1015 (C–O–C str), 749.75 (C–Cl str).
the reaction mixture, a mixed solution of propargyl alcohol
1H NMR (500 MHz, CDCl3): d 0.85 (t, 3H, –CH3), d 1.37 (q, 2H, (2 ml, 33.8 mmol), chloroform (4 ml) and anhydrous triethyl
–CH2–), d 3.39 (s, 6H, –CH2–O–), d 3.59 (m, 6H, –CH2–Cl), d 3.72 amine (2 ml, 14.34 mmol) were added drop wise over a period of
(m, 6H, –CH–CH2–O–), d 3.98 (m, 3H, HC–O–).
0.5 h under stirring. The resulting mixture was stirred at 0 ꢀC for
13C NMR (125 MHz, CDCl3): d 7.55 (–CH3), d 23.12 (–CH2–), d one hour followed by stirring at room temperature for 24 h. The
43.34 (–C–CH2–CH3), d 45.75 (–CH2–Cl), d 70.00 (–CH–OH), d product was ltered under vacuum, the lter cake was washed
72.12 (–CH2–O–), d 78.98 (–C–CH2Cl).
with THF and distilled water several times. The obtained CNT–
alkyne was dried under vacuum at 60 ꢀC for 24 h. The as
prepared CNT–COOH and CNT–alkyne were characterized by
using FT-IR, TGA and FE-SEM techniques.
2.4. Synthesis of azide terminated hyperbranched polyether
polyol (HBP–N3)
The above synthesized chlorinated hyperbranched polyol (5 g,
1 eq.) was dissolved in a solvent mixture of acetonitrile and
water (8 : 1). To this, excess sodium azide (7.2 g, 7.5 eq.) was
2.6. Procedure for click reaction between of HBP–N3 and
CNT–alkyne (CNT–HBP)
added portionꢀwise. The resulting reaction mixture was stirred 150 mg of CNT–alkyne was dispersed in 30 g of DMF in a round
for 36 h at 75 C. Aer completion of the reaction, the solvent bottom ask. The mixture was sonicated for 10 minutes and
was evaporated. The product was extracted in chloroform and then ushed with nitrogen. Then 27 mg of CuI, 100 mg HBP–N3
washed with brine solution for 2 times. The organic layer was and 0.02 g of DIPEA were added to the above solution. The
collected and dried over sodium sulfate. The product HBP–N3 reaction mixture was stirred under nitrogen atmosphere at
was obtained aer evaporating the solvent.
70 ꢀC for 12 hours. The reaction mixture was terminated by
FT-IR (KBr, cmÀ1): 3432.78 (O–H str), 2956.04 (asym –CH3 cooling the reaction ask in ice bath followed by exposure to air.
str), 2917.58 (asym –CH2 str), 2879.12 (sym –CH3 str), 2102.53 The reaction mixture was diluted with 10 ml DMF followed by
(N3 str), 1451.57 (C–O str), 1100–1015 (C–O–C str).
ultrasonication for 10 minutes. The resulting mixture was
1H NMR (500 MHz, CDCl3): d 0.82 (t, 3H, –CH3), d 1.31 (q, 2H, ltered and washed with THF, aqueous ammonium hydroxide
–CH2–), d 3.34 (m, 6H, –CH2–N3), d 3.45 (m, 6H, –C–CH2–O–), d solution and pure water thrice. The resulting product was dried
3.5–3.7 (m, 6H, –CH–CH–O–), d 3.92 (m, 3H, HC–OH).
13C NMR (125 MHz, CDCl3): d 7.39 (–CH3), d 23.12 (–CH2–), d
42.93 (–C–CH2–CH3), d 53.27 (–CH2–N3), d 65.52 (–CH2–O–), d
72.61 (–CH–OH), 78.86 (–CH2O–C–CH2–N3).
under vacuum to obtain CNT–HBP (Fig. 1).
2.7. Synthesis of polyurethane–CNT hybrid composites
(CNT–PU)
Calculated amount of CNT–HBP (0%, 0.5%, 1% and 2% with
respect to total weight) and H12-MDI (0.98 g, 3.73 mmol) were
2.5. Synthesis of alkyne terminated MWCNT (CNT–alkyne)
1.0 g of pristine CNT was treated with 23 ml of 50 wt% nitric taken in a round bottom ask. To this 2.5 g of cellosolve acetate
acid in a round bottom ask. The mixture was reuxed for 50 was added as a solvent. The mixture was sonicated for half an
hours. The resulting solution was diluted with a large amount of hour and stirred at 60 ꢀC for three hours. To this solution a
de-ionized water and ltered. The ltrate was washed several mixture of PTMG (1 g, 1 mmol) and TMP (0.15 g, 1.1 mmol, 15
times with de-ionised water until it reached a pH of 7, which weight% with respect to PTMG) dissolved in cellosolve acetate
ꢀ
was further dried at 120 C for 24 h. Finally, 0.91 g of carboxyl was added drop wise slowly. In all the hybrid composites, OH
functionalised CNT (CNT–COOH) was obtained. To the and NCO ratio was maintained as 1 : 1.2. The mixture was
Fig. 1 Synthetic protocol for hyperbranched polyol functionalized MWCNT (CNT–HBP) through azide–alkyne click chemistry.
24422 | RSC Adv., 2014, 4, 24420–24427
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 2 Synthesis of polyurethane–CNT hybrid composites.
ꢀ
stirred at 65 C for 8 hours under inert atmosphere. Different hyperbranched polymer. For this reason, rstly a chlorinated
CNT–PU hybrid composite lms were casted on a tin foil sup- hyperbranched polyether polyol (HBP–Cl) was prepared and
ported over a glass plate by a manual driven square applicator, which was converted into HBP–N3. The formation of HBP–Cl
with adding one drop of 5% DBTDL in MIBK as catalyst and one from Trimethylolpropane and epichlorohydrin is evident from
drop Tagostab as surfactant. The excess NCO present in the the disappearance of epoxide stretching frequencies around
hybrid lms were moisture cured at 30 ꢀC and laboratory 3050 and 900 cmÀ1 and the formation of ether stretching at
humidity condition (25–30%) for 15 days. The supported lms 1100–1000 cmÀ1 in FT-IR; the appearance of new peaks related
were extracted aer amalgamation and cleaning (Fig. 2).
to ether linkage at d 3.75, d 3.98 in 1H-NMR and d 72.21, d 70.13
in 13C-NMR; the presence of (M + Na) m/z peak at 435 in ESI-MS
spectra and also the absence of peaks related to epoxide at d 2.6,
d 2.8, d 3.2 in 1H-NMR and peaks at d 51.35, d 46.89 in 13C-NMR
simultaneously, is a substantial conrmation of the as prepared
HBP–Cl. Further, the nucleophilic substitution of HBP–Cl with
azide groups for the formation of HBP–N3 was corroborated
with the disappearance of C–Cl stretching at 750 cmÀ1 and the
appearance of sharp and intense peak at 2100 cmÀ1 in FT-IR
3. Results and discussion
3.1. Hyperbranched polyether polyol functionalised carbon
nanotube
The development of compatible hyperbranched polymer is very
essential for the functionalization of CNTs. This entails high
precision at various stages involved in the synthesis of the nal
Fig. 3 (a) TGA thermograms (b) Raman spectra of pristine CNT, CNT–COOH, CNT–alkyne and CNT–HBP; (c) TGA thermograms of 0%,0.5%, 1%
and 2% CNT–PUs. (d) The polarization curves of different CNT–PU coatings along with the bare metal evaluated by the Tafel method in a 5%
NaCl solution.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 24420–24427 | 24423

View Article Online
RSC Advances
Paper
1
studies; the disappearance of peaks at d 3.59 (–CH2–Cl) in H- ensure the solubility of CNTs and also facilitate the interaction
NMR and d 45.75 (–CH2–Cl) in 13C-NMR; simultaneously the between tubes through hydrogen bonding between the hyper-
formation of peaks at d 3.34 (–CH2–N3) in 1H-NMR and d 53.27 branched polyols. The so segments of HBP impart inherent
(–CH2–N3) in 13C-NMR; the presence of (M + Na) m/z peak at 454 amorphous nature to the CNT, which is observed as a decrease
1
in ESI-MS spectra. (For H-NMR, 13C-NMR, FT-IR and ESI-MS in intensity of the XRD peak.
spectra refer ESI†). Once HBP–N3 was synthesized the imme-
Raman analysis also clearly reveals the difference in each of
diate step involved is the modication of CNT surface. Since, the modied CNTs. The peak at 1312.42 cmÀ1 corresponds to
pristine CNTs cannot be used for the attachment of HBP–N3, it CNT and 1318.91 cmÀ1 for other functionalized CNT (Table S2†)
is obligatory that CNT undergoes certain changes. Thus CNTs corresponds to D-band (‘D’ for defect or disorder). From the
were initially oxidized to form CNT–COOH, then to CNT–alkyne experimental observations it is evident that there is a signicant
by treatment with propargyl alcohol. The click reaction between shi in D-band with modication of CNT surface. This is an
the azide terminated HBP–N3 and CNT–alkyne, conrmed by outcome of the disorders present and symmetry lowering. The
the disappearance of azide and alkyne stretching peaks around appearance of D-band in carbonaceous materials is because of
2100–2200 cmÀ1 in the FTIR spectra. Thermogravimetric anal- one-phonon second order Raman scattering process, which
yses were also performed in order to determine the changes in involves both elastic and inelastic scattering.33,34 Jorio et al.35
thermal stability of various samples. The overlay image of reported that the D-band shows a strong dependence on the
various TGA graphs is shown in Fig. 3(a). For instance, the onset excitation wavelength. The polarization dependence studies
decomposition temperatures of pristine CNT, CNT–COOH, carried out, suggest that origin of D-band is because of disorder
ꢀ
ꢀ
ꢀ
CNT–alkyne and CNT–HBP are 583.90 C, 572.14 C, 470.72 C and nite size effect of carbons. There is also evident change in
and 379.94 ꢀC respectively. The thermal stability of the CNT the ID/IG ratios in each modied CNT. ID/IG ratio provides
decreases with subsequent surface modication (Table S1†).
information on the alignment of CNT. G-band (G stands for
X-ray diffraction peaks of the CNT samples reveal two graphite) originates from rst order Raman effect. G-band
prominent peaks at 26.5ꢀ and 42ꢀ (Fig. S10†) which can be which comprises of six modes (according to group theoretical
attributed to the (002) and (100) plane of hexagonal graphite.31 analysis) for each symmetrical mode in the nanotubes, the
However there is a distinct change in the intensities of modied vibration of atoms are either along the tube axis (G+) or along
CNTs as compared to the pristine CNT. In the pristine CNT the the circumferential direction (GÀ). There is also a denite
van der Waal's force of attraction holds the nanotube bundles change in the intensities of G-band from the Fig. 3(b). The G-
together. Following the acid treatment, it is evidently seen in band in CNT corresponds to optical phonon mode between the
TEM and FE-SEM images that nanotubes become loosely two dissimilar carbon atoms in the unit cell. It is known to
bundled (Fig. 4 and S11†). This is credited to the decrease in the originate from the sp2 carbons present in the CNT.36
van der Wall's force of attraction between nanotubes which
In the present study the CNT surface is anchored with
leads to a reduced inter-tubular interaction as compared to hyperbranched polyol, so as to ensure the solubility of CNTs
pristine CNT.32 Since the aim of the work is to obtain CNTs that and also facilitate the interaction between tubes through
are soluble for use as composite materials, it is necessary to hydrogen bonding between the hyperbranched polyols. It is
have substantial interaction between nanotube and also its observed that the surface modied CNTs such as CNT–COOH,
immediate surroundings. Thus, in the present study CNT CNT–alkyne and CNT–HBP show improved dispersion stability
surface is anchored with a hyperbranched polyol, so as to in various solvents. Moreover, CNT–HBP shows higher solu-
bility behaviour because of its surface hydroxyl functional
groups. The presence of carboxyl, alkynyl and hydroxyl groups
facilitate swi dispersion of CNT–COOH, CNT–alkyne and
CNT–HBP in polar solvents because of their inherent polar
nature. This behaviour may be due to formation of hydrogen
bonds with solvent systems. The poor dispersion stability of the
modied CNTs in toluene solvent is accounted to the non-polar
nature of the solvent system. (Fig. S12 and Table S3†).
3.2. Functional CNT–HBP polyurethane–urea hybrid
composites (CNT–PUs)
Once the properties of CNT–HBP were established, the next step
is to covalently link it to the polyurethane matrix so as to
fabricate functional hybrid composites. The successful
synthesis of carbon nanotube incorporated-polyurethane
hybrids is based on the complete absence of free –NCO peak
around 2270 cmÀ1 in the FT-IR spectra (Fig. S5†), which indi-
cates complete moisture curing of the polyurethanes. The
Fig. 4 TEM images of (a) pristine CNT, (b) CNT–COOH, (c) CNT–
alkyne, (d) CNT–HBP (scale bar indicates 50 nm).
observed absorption bands (cmÀ1) at 3050–3700 (–NH stretch),
24424 | RSC Adv., 2014, 4, 24420–24427
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 5 AFM images of (a) 0% CNT–PU (b) 0.5% CNT–PU, (c) 1% CNT–PU (d) 2% CNT–PU.
2800–3000 (–CH stretch consisting of asymmetric –CH3 stretch
XRD analysis also conrm the increase in the crystallinity of
at 2957, asymmetric –CH2 stretch at 2920, symmetric –CH3 CNT–PUs with the CNT–HBP loading (Fig. S10†). From UTM
stretch at 2872 and symmetric –CH2 stretch at 2851 cmÀ1), analysis (Fig. S13† and Table 1), it is observed that both tensile
1600–1800 (–C]O stretching of amide I), 1500–1600 (amide II strength and elongation at break increases with CNT loading.
stretch consisting of a mixture of peaks dN–H, nC–N and nC–C), Since, CNTs are thermo-oxidative stable materials;38 their
1394 (dCH symm/assym), 1215–1350 cmÀ1 (amide III, nC–N), 766 incorporation enhances the thermal conductivity of poly-
2
(amide IV due to –NH out of plane vibration) and 720 (–CH2 urethane hybrid composites and therefore increases its thermal
rocking) supports the formation of polyurethanes. Neat poly- stability (Fig. 3(c) and Table 2). The reason for improvement of
urethane (0% CNT–PU) has a highly smooth surface, whereas thermal and mechanical stability of these hybrid lms are due
the incorporation of CNT transforms the PU lm into a rough to the formation of urethane linkages by the hydroxyl groups
and fractured micro structure with many pleats (Fig. S11†). The present on the surface of CNT–HBP. The formed urethane
effect of incorporating CNTs in the organization of PU matrix linkages are capable of cross-linking through hydrogen bond-
could be correlated with the increase in the density of pleats ings with the polyurethane matrix, which also imparts hydro-
with the percentage of CNT, which could be veried by the phobicity to the hybrids which is realized from water contact
surface roughness analysis from AFM (Fig. 5). The formation of angle measurements (Fig. S14†). Electrochemical polarization
these rough surfaces facilitates high elongation at break and curves (Fig. 3(d) and Table 3) from the Tafel plot suggests an
tensile strength of the polyurethane lm.37
improvement in the polarization resistance and deterioration in
the corrosion rate amongst different CNT–PU lms. This might
be due to the formation of more cross-linked structures with
increasing the CNT–HBP content, which restricts the penetra-
tion of corrosive species by acting as a strong physical barrier.
Thus, the cross-linking density of CNT–PU increases with
increase in the CNT percentile. Thus, CNT–PU lms are an
excellent barrier to atmospheric moisture and oxygen. Such
high cross-linking of CNT–HBP with polyurethanes instills
shape recovery property to CNT–PU hybrid materials, which
were examined by using UTM instrument in tensile mode. It is
Table
1 Mechanical properties of CNT–PU hybrids (from UTM
analysis)
Tensile strength
Elongation at
break (%)
Sample code
(N mmÀ2
)
0% CNT–PU
0.5% CNT–PU
1% CNT–PU
2% CNT–PU
1.25
2.031
4.31
6.25
10.74
40.5
136.5
146.3
Table 2 TGA profile of various CNT–PUs
Onset decomposition
10%, wt
50%, wt
% wt
Sample code
temperature (TON) (ꢀC)
loss temperature (Td10) (ꢀC)
loss temperature (Td50) (ꢀC)
remaining at 350 ꢀ
C
0% CNT–PU
0.5% CNT–PU
1% CNT–PU
2% CNT–PU
247.02
251.91
254.45
256.34
286.67
284.40
291.21
292.34
335.45
336.02
339.42
343.39
40.08
40.09
42.61
45.90
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 24420–24427 | 24425

View Article Online
RSC Advances
Table 3 The electrochemical parameters measured from the Tafel plots of different CNT–PU hybrid coatings along with bare metal
Paper
Coating thickness
(mm)
Polarization resistance
Corrosion rate
(CR) (mm per year)
Sample code
Ecorr (mV)
Icorr (nA cmÀ2
)
(Rp) (k Ohm cm2)
Bare mild steel panel
1% CNT–PU
2% CNT–PU
—
67
69
À476.6
À474.4
À200.0
3.86 Â 103
1.12 Â 103
164
8.85
490.64
3502.07
0.047
0.012
0.0019
Fig. 6 Shape recovery tests of 2% PUCNT film (a) manual and (b) UTM.
observed that, with increase of CNT loading, the shape recovery This mechanism restricts the generation of large recoverable
nature of the polyurethane lm increases. First the samples strain and imparts higher recoverable force levels. The stored
were elongated up to their break point at room temperature elastic strain energy in the particles is released once the lm is
using UTM. The elongated samples were removed from UTM unloaded. The temperature provided during the process help in
and kept at 60 ꢀC for recovery. The sample recovered its original rapid shape recovery. The transaction between recoverable
shape within 10 s of heat treatment and the process was displacement and force is an important characteristic of the
repeatable (Fig. 6 and Table 4). The shape recovery property of shape memory CNT–PUs. The lm with 2% loading was found
CNT–PU hybrids was attributed to the excellent mechanical to be optimum which showed a 100% shape recovery. CNT–PUs
stability of the CNT and high thermal conductivity coefficient. were also treated with different bacterial and fungal stains in
The shape recovery of CNT–PUs depends on the presence and order to estimate its tolerance to microbial attack. It is observed
volume fraction of the CNT–HBP in the polyurethane lm. that the lms with higher CNT–HBP content showed superior
Experimental studies reveal that the addition of CNT–HBP to PU anti-microbial properties (Fig. S15†).
lowers the unrestrained recoverable strain limit and increases
the attainable controlled recovery stress. The particles are
4. Conclusions
forced to store internal elastic strain energy during elongation.
The past decade has shown a remarkable improvement in the
carbon nanotechnology. However little has been researched
about polyurethane based hybrids. We have successfully
Table 4 Shape recovery behavior of various CNT–PU samples
synthesized multifunctional polyurethane–urea–CNT hybrid
100% shape recovery time
composites with superior mechanical, thermal, hydrophobic,
antimicrobial and anticorrosive properties along with excellent
shape recovery behavior. These properties are mainly due to
uniform distribution of decorated CNTs in polyurethane–urea–
CNT hybrid composite matrix as well as improved cross-linking.
This improvement is due to the presence of extra hydroxy
(approximately)
Sample code
% elongation
Room temp.
At 60 ꢀC
0.5% CNT–PU
1% CNT–PU
2% CNT–PU
30
130
130
40 s
25 s
20 s
20 s
10 s
10 s
24426 | RSC Adv., 2014, 4, 24420–24427
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
groups on decorated CNTs as compared to neat polyurethane. 15 J. N. Coleman, A. B. Dalton, S. Curran, A. Rubio, A. P. Davey
Though the research is fundamental, it can be exploited for and A. Drury, Adv. Mater., 2000, 12, 213–216.
applications as smart coatings for different substrates such as 16 P. Singh, S. Campidelli, S. Giordani, D. Bonifazi, A. Bianco
metals, textiles, leather, ber etc. The shape recovery with and M. Prato, Chem. Soc. Rev., 2009, 38, 2214–2230.
stimuli-responsive nature along with hydrophobicity can be 17 J. Feng, J. Sui, W. Cai and Z. Gao, J. Compos. Mater., 2008, 42,
exploited for smart wrinkle resistant textiles whereas the 1587–1595.
moisture curing, corrosion resistance, antimicrobial properties 18 V. C. Moore, M. S. Strano, E. H. Haroz, R. H. Hauge,
and hydrophobicity can be exploited for smart high perfor-
mance coatings on metals in moist and marine environments.
R. E. Smalley, J. Schmidt and Y. Talmon, Nano Lett., 2003,
3, 1379–1382.
Thus polyurethane–urea–CNT hybrid composites could repre- 19 O. Matarredona, H. Rhoads, Z. Li, J. H. Harwell, L. Balzano
sent a new exciting direction that may open up opportunities in
and D. E. Resasco, J. Phys. Chem. B, 2003, 107, 13357–13367.
various scientic elds.
20 M. J. O'Connell, P. Boul, L. M. Ericson, C. Huffman, Y. Wang,
E. Haroz and R. E. Smalley, Chem. Phys. Lett., 2001, 342, 265–
271.
Acknowledgements
21 C. A. Dyke and J. M. Tour, J. Phys. Chem. A, 2004, 108, 11151–
The present research was supported by CSIR under the Intel-
Coat Project (CSC-0114). The authors would like to extend their
thanks to the Director, CSIR-IICT for permitting the work to be
carried out.
11159.
22 S. Bose, R. A. Khare and P. Moldenaers, Polymer, 2010, 51,
975–993.
23 R. G. Mendes, A. Bachmatiuk, B. Buchner, G. Cuniberti and
¨
¨
M. H. Rummeli, J. Mater. Chem. B, 2013, 1, 401–428.
24 M. Kumar and Y. Ando, J. Nanosci. Nanotechnol., 2010, 10,
3739–3758.
25 J. T. Sun, C. Y. Hong and C. Y. Pan, Polym. Chem., 2011, 2,
998–1007.
26 A. M. Caminade and J. P. Majoral, Chem. Soc. Rev., 2010, 39,
2034–2047.
References
1 S. Iijima, Nature, 1991, 354, 56–58.
2 J. P. Salvetat, J. M. Bonard, N. H. Thomson, A. J. Kulik,
L. Forro, W. Benoit and L. Zuppiroli, Appl. Phys. A: Mater.
Sci. Process., 1999, 69, 255–260.
3 X. Wan and J. Dong, Phys. Rev. B: Condens. Matter, 1998, 58,
6756–6759.
27 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless,
Angew. Chem., 2002, 114, 2708–2711.
4 (a) D. Tasis, N. Tagmatarchis, A. Bianco and M. Prato, Chem.
Rev., 2006, 106, 1105–1136; (b) T. Fukumaru, T. Fujigaya and
N. Nakashima, Macromolecules, 2013, 46, 4034–4040.
5 M. Ouyang, J. L. Huang and C. M. Lieber, Acc. Chem. Res.,
2002, 35, 1018–1025.
28 W. H. Binder and R. Sachsenhofer, Macromol. Rapid
Commun., 2007, 28, 15–54.
´
29 S. Campidelli, B. Ballesteros, A. Filoramo, D. D. Dıaz, G. De
la Torre, T. Torres and J. P. Bourgoin, J. Am. Chem. Soc., 2008,
130, 11503–11509.
6 D. Qian, E. C. Dickey, R. Andrews and T. Rantell, Appl. Phys.
Lett., 2000, 76, 2868–2870.
7 M. J. Biercuk, M. C. Llaguno, M. Radosavljevic, J. K. Hyun,
A. T. Johnson and J. E. Fischer, Appl. Phys. Lett., 2002, 80,
2767–2769.
30 H. P. Price, US Pat., US3033803, 1962.
31 W. Li, C. Liang, W. Zhou, J. Qiu, Z. Zhou, G. Sun and Q. Xin,
J. Phys. Chem. B, 2003, 107, 6292–6299.
32 Q. Shi, D. Yang, Y. Su, J. Li, Z. Jiang, Y. Jiang and W. Yuan,
J. Nanopart. Res., 2007, 9, 1205–1210.
8 M. Cadek, J. N. Coleman, V. Barron, K. Hedicke and
W. J. Blau, Appl. Phys. Lett., 2002, 81, 5123–5125.
9 (a) C. Velasco-Santos, A. L. Martinez-Hernandez, F. T. Fisher,
R. Ruoff and V. M. Castano, Chem. Mater., 2003, 15, 4470–
4475; (b) L. C. Tang, Y. J. Wan, K. Peng, Y. B. Pei, L. B. Wu,
L. M. Chen and G. Q. Lai, Composites, Part A, 2013, 45, 95;
(c) L. C. Tang, H. Zhang, J. H. Han, X. P. Wu and Z. Zhang,
Compos. Sci. Technol., 2011, 72, 7–13.
33 R. Saito, A. Jorio, A. G. S. Filho, G. Dresselhaus,
M. S. Dresselhaus and M. A. Pimenta, Phys. Rev. Lett.,
2001, 88, 027401.
34 S. Osswald, M. Havel and Y. Gogotsi, J. Raman Spectrosc.,
2007, 38, 728–736.
35 A. Jorio, G. Dresselhaus, M. S. Dresselhaus, M. Souza,
M. S. Dantas, M. A. Pimenta and H. M. Cheng, Phys. Rev.
Lett., 2000, 85, 2617–2620.
10 A. Lendlein and R. Langer, Science, 2002, 296, 1673–1676.
11 B. S. Lee, B. C. Chun, Y. C. Chung, K. I. Sul and J. W. Cho,
Macromolecules, 2001, 34, 6431–6437.
36 H. M. Heise, R. Kuckuk, A. K. Ojha, A. Srivastava,
V. Srivastava and B. P. Asthana, J. Raman Spectrosc., 2009,
40, 344–353.
12 M. Wang, X. Luo and D. Ma, Eur. Polym. J., 1998, 34, 1–5.
13 L. S. Schadler, S. C. Giannaris and P. M. Ajayan, Appl. Phys.
Lett., 1998, 73, 3842–3844.
37 J. Huang, H. Zhang and Q. Zhang, J. Wuhan Univ. Technol.,
Mater. Sci. Ed., 2012, 27, 608–614.
38 H. Xia and M. Song, So Matter, 2005, 1, 386–394.
14 P. Liu, Eur. Polym. J., 2005, 41, 2693–2703.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 24420–24427 | 24427