See discussions, stats, and author profiles for this publication at: [Link]

net/publication/331680697

Efficacy of ultra-low loading of amine functionalized graphene oxide into

glycidol-terminated polyurethane for high-performance composite material

Article  in  Reactive and Functional Polymers · March 2019

DOI: 10.1016/[Link].2019.03.008

CITATIONS READS

0 113

3 authors:

Madhab Bera Pragya Gupta

Indian Institute of Technology Roorkee Indian Institute of Technology Roorkee
10 PUBLICATIONS   37 CITATIONS    8 PUBLICATIONS   19 CITATIONS   

SEE PROFILE SEE PROFILE

Pradip K. Maji
Indian Institute of Technology Roorkee
44 PUBLICATIONS   413 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

polymer composite for packaging application View project

Cellulose View project

All content following this page was uploaded by Madhab Bera on 22 March 2019.

The user has requested enhancement of the downloaded file.

 Reactive and Functional Polymers 139 (2019) 60–74

Contents lists available at ScienceDirect

Reactive and Functional Polymers

journal homepage: [Link]/locate/react

Efficacy of ultra-low loading of amine functionalized graphene oxide into T

glycidol-terminated polyurethane for high-performance composite material
Madhab Bera, Pragya Gupta, Pradip K. Maji
⁎

Department of Polymer and Process Engineering, Indian Institute of Technology Roorkee, Saharanpur Campus, Saharanpur 247001, UP, India

ARTICLE INFO ABSTRACT

Keywords: In this work, the effect of amine functionalization of graphene oxide on the physicomechanical properties of
Glycidol terminated polyurethane (GPU) glycidol-terminated polyurethane (GPU) nanocomposites was studied. p-phenylenediamine (PPD) functionalized
p-phenylenediamine modified graphene oxide GO (GO-PPD) was used for the fabrication of nanocomposites with GPU by an in-situ polymerization method.
(GO-PPD) Covalent functionalization of GO was confirmed by Fourier transform infrared spectroscopy (FTIR) and X-ray
Thermal and mechanical properties
photoelectron spectroscopy (XPS). Filler content of the nanocomposites was varied from 0.05 wt% to 0.2 wt%
and found the best result at 0.10 wt% loading. With the incorporation of just 0.10 wt% GO-PPD, tensile strength
of glycidol terminated polyurethane (GPU) was increased by 123%, Young's modulus was found to increase by
158% and elongation at break was increased by 27%. The maximum improvement in thermal stability as ob-
served from thermogravimetric analysis (TGA) was about 16 °C. The experimental result obtained with 0.10 wt%
loading of GO-PPD was compared with the same loading of GO and found that GO-PPD provides better im-
provement in both the thermal and mechanical properties of GPU compared to GO at the similar loading. Ultra-
low loading of modified GO has a tremendous influence in physicomechanical properties of polyurethane na-
nocomposites. Structure-morphology relationship was established through field emission scanning electron
microscopy (FESEM) and transmission electron microscopy (TEM).

1. Introduction polyurethane by incorporating nanofillers [4–11]. Bera et al. [4] re-

ported the effect of graphene oxide (GO) loading on the thermal, me-
Polyurethane (PU) is one of the most versatile macromolecular chanical and surface properties of PU. Maximum improvement in ten-
compounds having great technological importance. It usually consists sile strength and Young's modulus of the nanocomposite was 280% and
of a polyol, diisocyanate or polyisocyanate and a chain extender [1]. 64% respectively at 0.10 wt% loading of GO. Thermal stability of the
The block copolymer, PU has hard and soft segments where the hard nanocomposite also increases by 10 °C with the addition of 0.20 wt%
segment is responsible for its stiffness and strength and the soft segment GO. Pokharel et al. [12] also investigated that with the incorporation of
is responsible for flexibility. Generally, the percentage of hard and soft 1 wt% GO, tensile strength and elasticity of PU increases by 40.5% and
segment and H-bonding between urethane linkages determines the 19% respectively. Sadasivuni et al. [13] investigated the effect of hy-
physical and mechanical properties of PU [1,2]. Its properties can be drophilic graphene oxide (GO) and/or hydrophobically modified gra-
altered easily by changing ratio of one or more constituents. It can be phene oxide (mGO) sheets on dynamic mechanical, dielectric and
converted easily as per the demand of current and future technologies rheological properties of reinforced PU by varying the filler content
like composites, coatings, fibers, adhesives, etc. [3]. It has excellent from 0 to 3 wt%. Better properties were observed for modified GO filled
low-temperature properties, high impact resistance and very good PU composites at higher loading. Wang et al. [14] reported 40 °C im-
chemical and environmental resistance [2]. In spite of all these ad- provement in thermal stability of PU with the addition of 2 wt% func-
vantages, it has certain limitations like low stiffness and inadequate tionalized GO. Various blending approaches have also been adopted to
thermal stability. These can be improved in two ways: either by in- improve the stiffness and thermal stability of PU. One of the most im-
corporating nanofillers into the polyurethane matrix [4,5] or by mod- portant ones is the blending of thermoplastic polyurethane (TPU) with
ifying the polyurethane chain by reactive functionalization or blending epoxy resin. With the incorporation of epoxy resins, thermal stability
with another polymer [6]. There is already a large number of literature and mechanical properties of polyurethane increases and decreases its
available on the improvement of thermal and mechanical properties of combustibility. Kalita et al. [15] reported that the thermal and

⁎
Corresponding author.
E-mail address: [Link]@[Link] (P.K. Maji).

[Link]
Received 4 December 2018; Received in revised form 18 February 2019; Accepted 6 March 2019
Available online 12 March 2019
1381-5148/ © 2019 Elsevier B.V. All rights reserved.
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

mechanical properties of hyperbranched polyurethane could be im- After the completion of the reaction, the product was vacuum filtered
proved by blending with a different amount of bisphenol A based epoxy through a 0.2 μm nylon membrane filter. The residue was dispersed in
resin. However, in case of physical blending, there is always an issue of ethanol, ultrasonicated for 5 min and then vacuum filtered through the
compatibility between the two polymers which may lead to phase se- nylon membrane. The process of dispersion and filtration was carried
paration and hence the expected properties may not be achieved. Hsia out 5–6 times to remove the unreacted and excess PPD. The resultant
et al. [16] reported that glycidol termination in polyurethane leads to material was dried in an oven at 60 °C for 24 h [20,21] and 252 mg of
an increase in storage stability of polyurethane by converting iso- dried GO-PPD obtained. The percentage of PPD in GO-PPD is around
cyanate-terminated prepolymer to a glycidol terminated polyurethane 60%.
prepolymer. They also observed that polyurethane has poor compat-
ibility with commercially available epoxy resins. Yeganeh et al. [17] 2.3. Synthesis of GPU/GO-PPD nanocomposites
reported that mechanical and biodegradation properties of poly-
urethane improves after epoxy modification. Different combinations of Nanocomposites of GO-PPD and glycidol terminated polyurethane
poly(ε-caprolactone) based polyurethane and poly(ethylene glycol) were synthesized by an in-situ polymerization method. Briefly, 10 g
based polyurethane was used to get an optimum combination of bio- PTMEG and 10 ml dried DMF were taken in 250 ml 3- neck flask. The
degradation and mechanical property. With the amalgamation of two flask was placed in an oil bath and a water-cooled condenser was fitted
seemingly different technological branches; nanotechnology and to it. The materials were slowly heated up to 65 °C. 2.63 g MDI was
polymer technology, scientists are able to develop some high-perfor- injected into the flask after dissolving in DMF. The reaction was carried
mance composite materials that can bring revolution in the existing out for 2 h at 65 °C temperature in an inert (nitrogen) atmosphere. The
technology. Hussein et al. [18] investigated the positive improvement reaction medium was cooled down to 40 °C. 0.74 g glycidol was dis-
in mechanical properties of epoxy resin with the addition of GO-PPD. solved in 10 ml dried DMF and injected into the reaction mixture by
From the literature survey, it was clear that although there are various using a syringe. 3 drops of DBTL was also added separately and the
reports on graphene based polyurethane nanocomposites, epoxy mod- reaction temperature was raised slowly up to 75 °C. Stirring and heating
ification of polyurethane by using glycidol or epoxy resin, but there is were continued for another 2 h at the same temperature to synthesize
no report about the effect of PPD modified GO (GO-PPD) on the thermal glycidol terminated polyurethane. Required amount of GO-PPD was
and mechanical properties of glycidol-terminated polyurethane nano- dispersed in DMF and added to the glycidol terminated polyurethane.
composites. The reaction was continued for another 2 h at 75 °C. DMF was added
In this work, polyurethane backbone was modified by in-situ reac- occasionally to maintain reaction viscosity and stirring. Glycidol ter-
tion of isocyanate terminated polyurethane prepolymer with glycidol. mination and completion of GPU synthesis were confirmed by the
Thus, isocyanate-terminated polyurethane prepolymer is converted to presence of a small band of epoxy ring at 912 cm−1 in the FTIR spec-
glycidol terminated polyurethane prepolymer. This process in- trum. Also, the absence of any band at 2280 cm−1 confirmed that no
corporates epoxy groups within the polyurethane chain and the mod- isocyanate was present in excess (Fig. S1, Supporting Information).
ified system exhibits some advantageous properties like better thermal After the completion of the reaction, the temperature was cooled down
stability and superior mechanical properties compared to normal to 50 °C and a stoichiometric amount of curing agent (HMD) was added
polyurethane. Moreover, for better interaction between GPU with GO, and mixed for 5 min.
GO has been modified with p-phenylenediamine (PPD). The thermal The material was cast into a petri dish and kept at 50 °C for 3 days
and mechanical properties of the GPU/GO-PPD nanocomposites have first then at 90 °C for 1 day to ensure complete removal of DMF. The
been compared with the GPU/GO nanocomposite at ultra-low loading percentage of hard and soft segments in the synthesized GPU were
of filler and found significant positive effect of PPD modification on the 25.2% and 74.8% respectively. Samples with 0, 0.05, 0.10, and 0.20 wt
said properties. % GO-PPD were designated as GPU-0, GPU/GO-PPD 0.05, GPU/ GO-
PPD 0.10, and GPU/ GO-PPD 0.20 respectively. Stepwise chemical re-
2. Experimental actions associated with the synthesis of GPU is presented in Fig. 1 and
the chemical bonding between GO-PPD and GPU as well as GPU and
2.1. Materials HMD are represented in Fig. 2.

4,4′-methylene diphenyl diisocyanate (4,4’-MDI), poly (tetra- 2.4. Synthesis of GPU/GO nanocomposite
methylene ether) glycol (PTMEG) with Mn= 2000, 2,3-Epoxy-1-pro-
panol (glycidol), p-phenylenediamine (PPD) and graphite flakes GPU/GO nanocomposite with 0.10 wt% loading of GO was prepared
(~150 μm size) were purchased from Sigma-Aldrich, India. N-dimethyl by the same procedure as described above. Only, 0.10 wt% GO dis-
formamide (DMF) and other necessary chemicals were purchased from persion in dried DMF was added in the second step of the reaction in-
Rankem India. Hexamethylene diamine (HMD) was purchased from Hi stead of 0.10 wt% GO-PPD. The resultant nanocomposite sample was
Media Laboratories Pvt. Ltd., India. The entrapped moisture in PTMEG designated as GPU/GO 0.10.
was removed by placing in a 70 °C vacuum oven for 24 h. DMF was
made moisture free by using 4 Å molecular sieves. All other chemicals 3. Characterization
were used without any further purification.
Powder XRD (Rigaku Ultima IV, Japan) with CuKα radiation
2.2. Functionalization of GO with PPD (λ = 1.54 Å), operated at a scanning rate of 4°/min within the dif-
fraction angle 4°–60°, was used to determine the crystal structure of GO
Amount of GO required for the functionalization was synthesized and GO-PPD. Interlayer spacing between the layers were calculated by
from graphite flakes following the procedure described in our earlier using Bragg's law (Eq. 1).
work [19]. For the synthesis of PPD functionalized GO, 200 mg of the
n = 2d Sin (1)
synthesized GO was first dispersed in 200 ml of distilled water by
overnight stirring followed by 2 h ultrasonication using a bath sonicator Here, n is the order of reflection (normally n = 1), λ is the wavelength
(100 W). 2 g PPD was added to the GO dispersion after dissolving in of X-ray and θ is the diffraction angle.
100 ml of distilled water at 70 °C for 1 h. Subsequently, 1.2 ml NH3 Chemical structure of GO and GO-PPD was analyzed by FTIR
solution (25% w/v) was added to the mixture and the reaction mixture spectrometer (PerkinElmer FTIR, UTAR spectrum Two) with KBr. IR
was refluxed at 95 °C temperature for 6 h under continuous stirring. spectra of GPU and its polymer nanocomposites with different wt% of

61
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 1. Schematic illustration showing the synthesis of Glycidol terminated polyurethane (GPU).

Fig. 2. (a) chemical bonding between GPU and GO-PPD (b) Cross-linking reaction between GPU and HMD.

62
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

GO-PPD were recorded by using the same instrument but in attenuated samples for each composition was tested and the average result was
total reflectance (ATR) mode. In all cases, samples were scanned at a reported.
scan rate of 4 cm−1 within the scan range 4000–500 cm−1. Dynamic mechanical thermal analysis (DMTA) of neat GPU and its
Elemental composition of GO and GO-PPD was determined by using nanocomposite films were executed by using Anton Paar rheometer,
PHI 5000 Versa Probe III X-Ray Photoelectron Spectroscopy (XPS), MCR-102 in DMTA tension mode at a constant frequency of 1 Hz. The
Physical Electronics, USA. rectangular samples used in this testing were approximately
Surface morphology of GO, GO-PPD, GPU, and its composites were 40 mm × 5 mm × 0.50 mm in dimensions. All the samples were first
done by using field emission scanning electron microscopy (MIRA3, cooled to −100 °C by liquid nitrogen and then were scanned a heating
TESCAN, USA) at an operating voltage of 10 kV. In case of GPU and its rate of 5 °C/min within the temperature range − 100 °C to 60 °C.
composites, the tensile fractured surface was used for the analysis.
Transmission electron microscopy (TEM, Tecnai G2 20S-TWIN, FEI 4. Results and discussions
Netherlands) was also used for morphological characterization GO, GO-
PPD and its nanocomposites. GO and GO-PPD samples were prepared 4.1. Characterization of GO-PPD
by drop casting the aqueous and ethanol dispersion (0.02 mg/ml in
each case) of GO and GO-PPD respectively on the 400 mesh carbon X-ray diffraction study gives a clear indication about the functio-
coated copper grid. The grids were dried in a 40 °C hot air oven and nalization of GO with PPD. XRD patterns of graphite, GO and GO-PPD
used for analysis. are shown in Fig. 3. The characteristic peak of graphite appeared at
The composite sample for TEM was prepared by cryo-ultra- 2θ = 26.5° corresponding to an interlayer spacing of 0.336 nm. Upon
mocrotomy with a Leica Ultracut UCT (Leica Microsystems GmbH, oxidation of graphite, the diffraction peak at 2θ = 26.5° completely
Vienna, Austria). 120 nm thick cryosection was prepared by cutting the disappeared and a new peak at 2θ = 9.4° appeared in GO [Fig. 3b]. The
sample at −80 °C temperature using a freshly sharpened glass knife interlayer spacing corresponding to the peak is 0.94 nm which is quite
with a cutting edge of 45°. The samples were then transferred to 400 higher than the interlayer spacing in graphite (~0.34 nm). This clearly
mesh carbon-coated Cu TEM grids and used for analysis. indicates the introduction of oxygen-containing functional groups in
Atomic force microscopy (Model: Dimension Icon, Bruker, USA) was the basic carbon structure [19]. The X-ray diffraction peak of GO is
used for the morphological characterization of GO and GO-PPD. AFM shifted to 2θ = 5.8° after reaction with PPD (i.e. in GO-PPD). Interlayer
samples were prepared by drop casting a very dilute solution (0.02 mg/ spacing corresponding to this peak is 1.52 nm which is again higher
ml) of GO and GO-PPD on glass slide followed by evaporation of the than that of GO (d001=0.94 nm). This observation clearly confirmed the
solvent. AFM images were captured in Scan Asyst mode at 512 lines per chemical functionalization of GO surface with PPD [20,22]. It addition
sample. Resulting images were analyzed by using NanoScope analysis to the diffraction peak at 2θ = 5.8°, GO-PPD also exhibit a broad band
software, version1.4 to get the size and thickness of the nanosheets. centered at 2θ = 24°. This broad band is due to partial reduction of GO
Tensile properties of glycidol terminated polyurethane and its na- during the functionalization with PPD [20]. Zhong et al. [22] and Ma
nocomposites with different wt% of GO-PPD were investigated by using et al. [20] also observed a similar type of XRD pattern for GO-PPD. It
the Instron Universal Testing machine (Model No 3365, Instron Co., was further confirmed by FTIR and XPS analysis.
Macclesfield, UK). The testing was performed as per ASTM D 412 Chemical functionalization and partial reduction of GO with PPD
standard, at a crosshead speed of 100 mm/min and using 5 kN load cell. was studied by using FTIR spectroscopy as shown in Fig. 3b. In the FTIR
The average result of five measurements for each composition was re- spectra of GO, a broad peak was observed at 3432 cm−1 due to OeH
ported. stretching vibration of carboxylic or hydroxyl groups present in GO
Thermal transitions of all the polymeric samples were measured in [19]. In addition to this, a few more significant peaks were observed in
differential scanning calorimeter (DSC 25, TA instrument, USA). The GO at 1730, 1630, 1420, 1224 and 1057 cm−1 corresponding to C]O
temperature ramp was carried out at 10 °C/min under an inert atmo- stretching vibration of carbonyl/carboxylic group, C]C stretching vi-
sphere maintained by a constant flow of nitrogen at 40 ml/min. bration of aromatic rings, OeH bending vibration, CeO stretching vi-
Thermogravimetric analysis (TGA) of GO, GO-PPD, GPU and its bration of alkoxy group and CeOeC stretching vibration of epoxy
nanocomposites was carried out in an EXSTAR TG/DTA 6300 (SII group respectively [19,23]. After reaction with PPD (GO-PPD), a new
NanoTechnology Inc.) instrument. For every measurement, about sharp peak appeared at 3453 cm−1. This could be due to the overlap of
10 mg sample was placed into an alumina pan and heated up to 800 °C OeH stretching peak with the NeH stretching. It will be more clear
at a heating rate of 10 °C/min under a nitrogen atmosphere. Three from the deconvoluted C 1 s XPS spectra (Fig. 4b) where OeH peak is

Fig. 3. (a) XRD pattern of graphite, GO and GO-PPD; (b) FTIR spectra of PPD, GO and GO-PPD.

63
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 4. (a) XPS survey spectra of GO and GO-PPD; (b, c) C 1 s XPS spectra of GO and GO-PPD respectively.

also visible in addition to the newly incorporated CeN peak. FTIR a new peak appeared at 285.6 eV corresponding to CeN group of GO-
spectrum of GO-PPD also contains a weak peak at 1724 cm−1 due to the PPD [24]. The appearance of CeN peak in the XPS spectrum of GO-PPD
C]O stretching vibration of unreacted carbonyl groups. The functio- confirms the functionalization of GO with PPD via SN2 reaction be-
nalization and grafting of PPD in GO were confirmed by the appearance tween the epoxy group of GO and the NH2 group of PPD. The observed
of new peaks at 1642 cm−1 (C]O stretching of amide group), results were in accordance with the available literature [20,24].
1568 cm−1 (NeH bending of amide group), 1190 cm−1 (CeN Morphological characterization of GO and GO-PPD was done by
stretching vibration), 812 cm−1 (NeH wagging of amide group) [20]. using field emission scanning electron microscopy (FESEM), atomic
These results confirmed the formation of amide linkage in GO-PPD and form microscopy (AFM) and transmission electron microscopy (TEM)
the results are consistent with the earlier reported values [20–22]. and shown in Fig. 5 Fig. S2 (Supporting Information) and Fig. 6 re-
XPS was used to investigate the changes in chemical structure oc- spectively. From the FESEM images (Fig. 5), it is clear that the size of
curred due to functionalization of GO with PPD. XPS survey spectra of GO sheets decreases remarkably after PPD functionalization. This is
GO and GO-PPD is shown in Fig. 4a and the C 1 s XPS spectra of GO and either due to the reduction of GO or due to the effect of strong amine
GO-PPD is represented in Fig. 4b and Fig. 4c respectively. Elemental (PPD) which causes a structural breakdown. In the FESEM image of GO,
analysis results showed that GO-PPD contains 10.22% nitrogen (N) the nanosheets are folded, crumpled and closely associated with each
which was absent in GO and the carbon to oxygen (C/O) ratio was other.
increased from 1.5 for GO to 5.8 for GO-PPD. This observation clearly The strong π-π interaction between the nanosheets is responsible for
confirmed that the nitrogen-containing functional group has been at- such observation [25]. Highly exfoliated morphology of GO was ob-
tached to GO surface at the same time partial reduction of GO occurred served for freeze-dried material. This kind of exfoliated morphology of
during functionalization [20]. Similar information was obtained earlier GO was observed in our previous work also [19]. The morphology
from the XRD analysis also. observed in this report is common for GO and GO-PPD and similar type
In order to get detailed information about the surface functional of morphology has been reported by other researchers [23,26]. The
groups present in GO and GO-PPD, deconvolution of C 1 s spectra was atomic percent of respective elements present in GO and GO-PPD was
done. In GO, four different peaks emerged at 284.8, 286.6, 287.6, and obtained from the Energy Dispersive X-Ray Analysis (EDAX) of FESEM
288.8 eV corresponding to CeC/C]C (unoxidized skeletal carbon), images as presented in Fig. 5 (c, d). From the EDAX analysis, it is clear
CeOH (hydroxyl group), CeOeC (epoxy group) and C]O (carbonyl that nitrogen (N) was present in GO-PPD which was confirmed earlier
group) respectively [20,24]. By comparing Fig. 4b and Fig. 4c, it is clear by XPS analysis also. Moreover, the increase in C/O from 1.6 to 4.8 was
that the intensity of CeOH and C]O peak decreased significantly after also observed after functionalization of GO with PPD which proves the
reaction of GO with PPD. This indicates reduction of GO during func- reduction of GO during functionalization. This result is consistent with
tionalization. Also, the CeOeC peak at 287.6 eV in GO disappeared and the XPS study, discussed previously.

64
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 5. (a, b) FESEM images of GO and GO-PPD respectively; (c, d) EDAX spectra of GO and GO-PPD respectively (elemental percent. Were shown inside the
respective image).

AFM also provides similar morphological information as obtained in 0.90 nm for GO and 3.1 nm for GO-PPD. The thickness of GO was little
FESEM and TEM analysis. The size reduction of GO sheets are also higher than the single layer graphene (0.34 nm) indicating the presence
visible in AFM images (Fig. S2). Average size of GO sheet as obtained of oxygen containing functional groups in GO surface. Moreover,
from AFM analysis was around 6.0 μm and the same for GO-PPD was average thickness of GO-PPD was quite higher than GO. This clearly
1.5 μm. In addition to this, sectional analysis of Scan Assist mode AFM indicates the attachment of PPD in GO surface and causes an increase in
image provides the average thickness of the nanosheets which was sheet thickness.

Fig. 6. TEM images of (a) GO and (b) GO-PPD, (SAED patterns were shown in the inset).

65
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

TEM is another electron microscopic technique used for direct improvement in thermal stability of GO-PPD over the GO is due to the
morphological characterization of GO and GO-PPD. The bright field incorporation of stable PPD and reduction of GO during functionali-
TEM images of GO and GO-PPD are presented in Fig. 6. As seen in zation [20]. Yuan et al. [24] reported a similar type of results for GO
Fig. 6a, GO nanosheets were exfoliated, wrinkled and transparent. The and GO-PPD used in functionalized graphene/polypropylene nano-
size of the GO nanosheets was about a few micrometers. The wrinkled composites.
topography was mainly due to the folding of extremely thin and long
size GO nanosheets and also due to the existence of oxygen-containing
4.2. Structural characterization of GPU and its nanocomposites
functional groups on the GO surface [27]. The morphology of GO-PPD
was quite different from GO. Stacked layered structure with a smaller
FTIR spectroscopy was used to verify the successful synthesis of GPU
size of the flake was observed for GO-PPD as seen in FESEM also. In
and its nanocomposites (Fig. 8). It also provides valuable information
Fig. 6b, the dark area indicates the stacking of several layers of GO-PPD
regarding the chemical interactions between GPU/GO-PPD and GPU/
and the lighter area signifies less or no stacking [28]. By comparing the
GO. FTIR measurement of all samples had been carried out in atte-
TEM images it was clear that the basic layered structure of GO was
nuated total reflectance (ATR) mode. Some characteristics peaks were
retained even after functionalization with PPD. Selected area electron
identified for GPU and its nanocomposites. The bands at
diffraction (SAED) patterns were also seen on the GO and GO-PPD
3301–3313 cm−1 represents the stretching vibrations of H-bonded
sheets and the results are presented in the inset image of Fig. 6. The
NeH groups. Peaks at 1732 cm−1 (free C]O stretching), 1708 cm−1
bright diffraction spot in the SAED pattern of GO clearly indicates the
(H-bonded C]O stretching), 1539 cm−1 (CeN stretching), 2854 cm−1
presence of crystalline order in the structure of GO [19,29]. Whereas, in
and 2940 cm−1 (symmetric and asymmetric stretching of CH2) are also
case of GO-PPD no such bright spot observed confirming the absence of
important and confirms the synthesis of urethane linkage [4]. Absence
crystalline order in GO-PPD structure. This observation complies with
of any band at ~2280 cm−1 indicates that no isocyanate was remained
the results found from XRD.
in excess in all samples [7,32]. Appearance of a weak band at 912 cm−1
Thermal stability of GO and GO-PPD was determined from ther-
for uncrosslinked GPU (Fig. 8A (a)) confirmed the incorporation of
mogravimetric analysis (TGA). TGA and its corresponding differential
epoxy ring within the polyurethane backbone [15]. Furthermore, the
curves are recorded in Fig. 7. From Fig. 7a it is quite clear that the
disappearance of this band in all samples, after cross linking confirmed
decomposition of GO-PPD occurs at a slower rate than GO. GO exhibits
that the epoxy ring was involved in curing reaction. The same is shown
three-step degradation in nitrogen atmosphere. The first step degrada-
more clearly in Fig. 8C. From Fig. 8B, it is clearly visualized that the
tion is due to the removal of entrapped moisture within the GO nano
NeH stretching band was shifted towards lower value with the addition
sheets and the second step degradation is due to the removal of labile
of GO-PPD/GO and the change become maximum in case of 0.10 wt%
oxygen containing functional groups from the GO surface. The sharp
GO-PPD loaded composite. This observation clearly confirmed that the
degradation of GO at around 176 °C is due to the rapid volume ex-
H-atom of NeH bond was involved in H-bonding or any other kind of
pansion of GO at higher heating rate. The rapid volume expansion oc-
interaction with GO-PPD [4,18].
curs due to very fast removal of entrapped moisture present in GO
sheets, as well as removal of other volatile gases coming from the
thermal degradation of GO at high heating rate. According to Zhang 4.3. Chemical interaction between GPU and GO-PPD by XPS analysis
et al. [30] when the ramping speed in TGA become greater than 5 °C/
min then this kind of dramatic mass loss is observed for GO. Similar Formation of chemical bonds between the epoxy group of GPU and
type of rapid degradation of GO was also observed by Mondal et al. the NH2 group of the PPD functionalized GO was also confirmed by XPS
[31]. The final degradation occurs at around 500 °C due to the removal analysis. The deconvoluted C1s XPS spectra of GPU and GPU/GO-PPD
of stable oxygen containing functional groups [20]. The temperature for 0.10 without adding crosslinker is shown in Fig. 9. From the decon-
5% decomposition (Tin) or onset of decomposition for GO was 71.3 °C voluted curves (Fig. 9a and Fig. 9b), it was clear that the peak for epoxy
whereas the same for GO-PPD was 75.2 °C. Moreover, the temperature group was present in GPU; but, after adding 0.10 wt% GO-PPD the peak
corresponding to maximum decomposition temperature (Tmax) for GO for epoxy group was vanished and the peak area for the CeN increases.
was only 176 °C compared to 640 °C for GO-PPD (Fig. 7b). This observation clearly confirms that a chemical bond has been formed
So, GO-PPD was thermally more stable compared to GO. The huge between epoxy group of GPU and the NH2 group of GO-PPD. Similar
indication was also observed from FTIR analysis also.

Fig. 7. (a) TGA and (b) derivative weight loss curves of GO and GO-PPD.

66
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 8. (A) full range FTIR-ATR spectra; (B) NeH stretching bands; (C) magnified image of full range spectra within 1000–700 cm−1 for (a) uncrosslinked GPU; (b)
GPU-0; (c) GPU/GO-PPD 0.05; (d) GPU/GO-PPD 0.10; (e) GPU/GO-PPD 0.20; (f) GPU/GO 0.10.

4.4. Morphological characterization of GPU and its nanocomposites this context, morphological characterization of polymer nanocompo-
sites is very important. In the current study, morphological character-
In addition to the interfacial interaction between polymer and filler, ization was done by FESEM and TEM analysis to determine the dis-
dispersion of nanofiller within the polymer matrix plays an important persion state of nanofillers in polymer matrix.
role in determining the final properties of polymer nanocomposites. In Fig. 10(a-e) represents the FESEM image of the tensile fractured

Fig. 9. Deconvoluted C1s XPS spectra of GPU and GPU/GO-PPD 0.10 without crosslinker.

67
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 10. FESEM images of (a) pristine GPU; (b) GPU/GO-PPD 0.05; (c) GPU/GO-PPD 0.10; (d) GPU/GO-PPD 0.20; (e) GPU/GO 0.10. (f) TEM image of GPU/GO-PPD
0.10.

surface of GPU and its nanocomposites From Fig. 10a, it was quite clear The results observed in TEM support the previously observed result
that tensile fractured surface of neat GPU was relatively flat and from FESEM. Raghu et al. [36] reported the similar type of morphology
smooth. However, with the addition of GO-PPD, roughness of the for waterborne polyurethane/functionalized graphene nanocomposite.
fractured surface increases due to irregular protuberances on the sur- The TEM image suggests that the layered structure of GO-PPD still re-
face [4,33]. Dark area of the FESEM images represent the soft segment mained within the polymer nanocomposite.
of polyurethane matrix and the bright part of the images represent the
broken graphene nanosheets attached with the hard segment [34].
4.5. Mechanical properties of GPU and its nanocomposites
There are wrinkles and ripples in the FESEM images which is due to the
π-π interaction of the nanosheets and n-π interaction between hard
Mechanical properties of a polymer nanocomposite mainly depend
segment of GPU and nanosheets. At 0.10 wt% loading of GO-PPD, the
on two parameters. First, the dispersion of nanofiller within the
nanosheets were evenly distributed within the GPU matrix (Fig. 10c).
polymer matrix and second, the interfacial interaction between polymer
This is due to improved interactions (covalent and H-bonding) between
and filler [37]. Dispersion of GO-PPD within GPU has already shown in
the GPU and GO-PPD which leads to tearing of the nanosheets during
Fig. 10 and the chemical interaction between the PPD functionalize GO
tensile failure instead of complete pull out. At higher loading ag-
and epoxy ring of GPU has been proved from the FTIR-ATR and XPS
gregation of filler took place due to higher filler- filler interactions
analysis. The effect of GO-PPD loading in the mechanical properties of
compared to filler-matrix interaction [4] which results in irregular
GPU was determined from the stress-strain curve (Fig. 12a) and the
morphology (Fig. 10d).
results are depicted in Table 1. Tensile strength, Young's modulus and
In the case of GPU/GO 0.10 nanocomposite, GO was uniformly
elongation at break increases with GO-PPD loading and become max-
distributed within the matrix as seen in Fig. 10e but it was not so
imum at 0.10 wt% loading. Tensile strength of GPU was 19.69 MPa
uniform as GPU/GO-PPD 0.10 (Fig. 10c). This might be due to better
which is quite higher than the tensile strength of thermoplastic poly-
interaction between GPU and GO-PPD compared to GPU and GO. Bera
urethane without glycidol termination (8.5 MPa), reported in our pre-
et al. [4] reported the similar type of morphology for thermoplastic
vious work [4]. Also, there was a huge difference in Young's modulus
polyurethane/GO and RGO based polymer nanocomposite. This type of
value too (4.15 MPa for GPU and 2.30 MPa for TPU [4]). These differ-
morphology has been reported in many other literatures also [22,35].
ences arises from the crosslinking reaction between the epoxy rings by
TEM was used for direct morphological observation of nano-
the HMD crosslinker. Tensile strength of GPU/GO-PPD composites in-
composites. It was employed to visualize the dispersion state of GO-PPD
creases up to 0.10 wt% loading of GO-PPD and becomes the maximum
within the GPU matrix. The bright field TEM image of GPU/GO-PPD
at this loading. With further addition of GO-PPD, tensile strength de-
0.05, GPU/GO-PPD 0.10 GPU/GO-PPD 0.20 and GPU/GO 0.10 samples
creased. A substantial enhancement in tensile properties of the in-situ
are shown in Fig. 11. From the TEM image, it is clearly visible that GO-
generated polyurethane nanocomposites was due to the improved
PPD nanosheets are uniformly dispersed within the polymer matrix at
covalent interaction between -NH2 group of GO-PPD and the epoxy ring
0.10 wt% loading (Fig. 11b). With further increase in loading, the na-
and crosslinking reaction by HMD. Another possible reason is better
nosheets are aggregated within the polymer matrix which has been
dispersion of GO-PPD at 0.10 wt% loading as seen in Fig. 10 c &
shown in Fig. 11c by red dotted circles. This is due to greater interac-
Fig. 11b. However, at further higher loading filler-filler interaction
tion between the filler over the polymer-filler interactions. A relatively
dominates over the filler-polymer interaction leading to phase sepa-
different type of TEM morphology is observed in case of GO filled GPU
rated type morphology (Fig. 10d) and hence decrease in properties [4].
nanocomposite (Fig. 11d) where the lines represent the folded GO
Again, GO-PPD posses much higher modulus compared to polyurethane
sheets, which are dispersed into the GPU matrix [8]. The dark portion
so Young's modulus of the nanocomposites goes on increasing with
of the image indicates aggregation of GO. The folding arises due to
increase in filler loading up to 0.10 wt%. The decrease in modulus at
larger size of the GO nanosheets which is already seen from microscopic
0.20 wt% is due to the aggregation of GO-PPD which leads to inferior
analysis of the materials.
interaction between polymer and filler causing lower modulus value

68
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 11. TEM micrographs of (a) GPU/GO-PPD 0.05; (b) GPU/GO-PPD 0.10; (c) GPU/GO-PPD 0.20; (d) GPU/GO-PPD 0.10.

[38]. In order to verify the effect of PPD modification, GPU nano- clearly confirmed that GO-PPD has better interaction with GPU com-
composite sample was prepared by adding 0.10 wt% of unmodified GO. pared to GO. By comparing the results obtained from stress-strain curve
Interestingly, the GPU/GO 0.10 nanocomposite showed inferior me- of GPU/GO 0.10, it is be quite clear that although the mechanical
chanical properties compared to GPU/GO-PPD 0.10. This observation properties of GPU/GO 0.10 were inferior to GPU/GO-PPD 0.10, the

Fig. 12. (a) Stress-strain curve of GPU and its nanocomposites with different wt% GO-PPD; (b) Variation of tensile strength and Young's modulus with GO-PPD
loading.

69
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Table 1
Tensile test result of EPU and its nanocomposites with different wt% GO-PPD.
Sample ID Tensile strength (MPa) Young's modulus (MPa) Elongation at break (%)

GPU-0 19.69 ± 0.75 4.15 ± 0.13 526 ± 21

GPU/GO-PPD 0.05 25.20 ± 1.21 5.20 ± 0.19 597 ± 19
GPU/GO-PPD 0.10 43.92 ± 1.52 10.7 ± 0.32 664 ± 23
GPU/GO-PPD 0.20 33.02 ± 1.35 8.95 ± 0.43 535 ± 25
GPU/GO 0.10 36.88 ± 1.46 8.46 ± 0.39 627 ± 26

value was higher than GPU/GO-PPD 0.05 and GPU/GO-PPD 0.20. This the degradation of hard segment [35]. Temperature corresponding to
could be due to the larger size of GO compared to GO-PPD which in- 5% weight loss was considered as initial degradation (Tin) and the
creased its reinforcing ability through simple H-bonding interaction temperature corresponding to maximum rate of degradation is re-
with urethane linkage [7]. Elongation at break for all nanocomposites is presented as Tmax. From Table 2, it is quite understandable that both the
higher compared to pristine GPU. There exist two reasons behind such initial degradation temperature and the maximum rate of degradation
observation. First reason is the plasticization effect exerted by GO-PPD. temperature increase with increase in GO-PPD loading. Temperature
At very low loading (0.10 wt%) GO-PPD flakes plasticizes the GPU for maximum rate of degradation increased by 16 °C at 0.2 wt% loading
chains by standing between them and thus increases their free volume of GO-PPD. The increase in thermal stability can be explained from the
and phase separation. Second reason is the existence of sacrificial H- improved interaction between GPU and GO-PPD through chemical
bond between the GPU grafted GO-PPD and the adjacent GPU chains. bonding and other physical or chemical interactions [4]. Further, the
During tensile loading, the motions of GPU chains stimulate the T5% and Tmax value of nanocomposite filled with GO-PPD was higher
movement of grafted GO-PPD which is bonded to another GO-PPD via than that of GO filled composite at the similar loading. This might be
another GPU chain. Hence, most of the tensile load is transferred by the due to the fact that GO-PPD has greater thermal stability compared to
H-bonds present between the covalently bonded GPU chains. An addi- GO, discussed earlier. Also, the covalent bonding between epoxy group
tional increase in load leads to disruption of H-bonds instead of cova- of GPU and NH2 group of GO-PPD was strong enough compared to the
lent bond present in nanocomposites and slipping of GO-PPD layers H-bonging interaction between oxygen containing functional groups of
over one another, causing an increase in elongation at break of the GO and the urethane linkage of GPU. The H-bonding interaction was
nanocomposites compared to neat GPU. If the elongation at break value also present between GO-PPD and GPU. Further, GO-PPD has much
for GPU/GO-PPD 0.10 is compared with our previously published result higher thermal stability compared to GO as discussed previously.
for unmodified PU and GO system at 0.10 wt% loading [4], it is quite Overall, GO-PPD based nanocomposites showed enhanced thermal
obvious that elongation at break value decreases after modification. stability over GO-based composites at the similar loading [20].
This could be due to partial crosslinking in polyurethane chains which
restrict the molecular chain movement leading to a decrease in elon-
[Link]. Calculation of filler content from TGA analysis. Filler content of
gation at break and an increase in modulus and tensile strength.
a reinforced polymer nanocomposite can be determined from the
residue of thermogravimetric analysis and by using theoretical
4.6. Thermal properties equations. This gives an indication about the dispersion of filler
within the polymer matrix as well as the effectiveness of the
4.6.1. Thermogravimetric analysis (TGA) processing technique used. However, the method has a serious
Thermal stability of the GPU and its polymer nanocomposites was drawback. When the thermal degradation temperature of polymer
determined from thermogravimetric analysis. TGA and the corre- and filler is very close, the method is not able to determine the filler
sponding DTG thermograms of GPU and its nanocomposites under N2 content accurately [39]. In the current study, filler content of GO-PPD
atmosphere are presented in Fig. 13a and Fig. 13b respectively. Fig. 13a and GO reinforced GPU nanocomposites have been determined by the
reveals that GPU and its all nanocomposites show two step degradation procedure reported by Nabinejad et al. [39]. The temperature at which
[4,11,35]. First step degradation occurs within 320–440 °C and corre- the majority of thermal degradation of polymer occurred is denoted as
sponds to the degradation of soft segment of GPU. The second step Tdp.
degradations occur within 440–650 °C temperature corresponding to The mass loss at Tdp can be correlated with the mass loss of the

Fig. 13. TGA thermogram of EPU and its polymer nanocomposites with different wt% GO-PPD.

70
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Table 2
Thermal properties of GPU and its nanocomposites obtained from TGA and DSC.
Sample TGA DSC

Tin (°C) Tmax (°C) Tg (°C) Tc (°C) Tm (°C) ΔHm (J/g) χc (%)

GPU-0 311.1 408.3 −62.9 −15.1 25.8 37.1 21.5

GPU/GO-PPD 0.05 313.8 412.0 −61.4 −14.2 27.0 40.5 23.5
GPU/GO-PPD 0.10 319.3 420.4 −59.6 −12.4 27.7 48.1 27.9
GPU/GO-PPD 0.20 324.4 424.2 −60.1 −10.2 28.8 48.5 28.2
GPU/GO 0.10 318.5 418.0 −61.8 −10.4 27.4 47.4 27.5

composite (mdc) by Eq. (2). 4.6.2. Differential scanning calorimetry (DSC)

The influence of GO-PPD on the melting and crystallization beha-
mdc = Mdf × mf + Mdp × mp (2)
vior of GPU was investigated by DSC. DSC curves of GPU and its na-
Here, mdc is the total mass loss of the composite and Mdp and Mdf are the nocomposites with different wt% of GO-PPD are presented in Fig. 14
percentage decrease in mass of the polymer and filler respectively at and the results are tabulated in Table 2. From the DSC curves, it is clear
Tdp. that GPU doesn't have any melting peak for hard segment because of its
Mass residue at final temperature (here it is considered as 650°C lower percent (~20%) [4]. Glass transition temperature (Tg) is related
where most of the degradation occurs) can be expressed by Eq. (3). to the physical aging of materials and observed from the DSC first
heating curve (Fig. 14a). Tg of soft segment (SS) of GPU increases with
mrc = Mrf × mf + Mrp × mp (3) the addition of GO-PPD and the effect become maximum at 0.10 wt%
Here, mrc is the mass residue of the composite at final temperature (Tr); loading where a 3.3 °C increase in Tg was observed. This was due to the
Mrf and Mrp are the filler and polymer residue respectively in percentage restriction in molecular chain movement due the chemical bonding
at Tr. mf is the mass of dried filler and mp is the mass of polymer. Now, interaction between GPU and GO-PPD. Also, the H-bonding interaction
filler mass percentage can be calculated from Eq. (4). between the urethane linkage of GPU and the residual oxygen func-
tionalities play an important role in it. The crystallization behavior of
100
Pf = [ × mdc + × mrc] GPU also changed with GO-PPD loading. Crystallization temperature
mc (4) (Tc) obtained from the DSC cooling curve, showed that Tc increases with
Here, α is the mass drop coefficient and β is the mass residue coefficient GO-PPD loading due to strong chemical interaction. Melting tempera-
and can be expressed by Eq. (4a) and Eq. (4b) respectively. ture (Tm) of nanocomposite increases with GO-PPD loading due to im-
proved compatibility between polymer and filler, and also the restric-
Mrp tion of molecular chain movement [11]. Melting point increases from
= × 100
(Mrp × Mdf Mrf × Mdp ) (4a) 25.8 °C to 28.2 °C with the addition of just 0.20 wt% of GO-PPD to GPU.
Also, the melting enthalpy increases with increase in GO-PPD loading
Mdp
= × 100 which is a direct indication of the increase in degree of crystallinity of
(Mrp × Mdf Mrf × Mdp ) (4b) the materials. The degree of crystallinity (χc) was calculated by using
the Eq. (5)
First, we have calculated the Tdp value from the TGA and DTG
curves which is 408.3 °C. Then mass drop, mass residue, mass drop Hm
percentage of polymer, filler and composites has been determined from Xc = x100%
Hm0 (5)
the TGA curve at 408.3 °C temperature. Mass residue and mass residue
percentage of the polymer, filler and their nanocomposites are also where, ∆Hm is the enthalpy of melting of the soft segment of poly-
determined at the final temperature (Tr). The results obtained from the urethane and its composites. ∆Hm0 = 172.2 J/g is the theoretical value
TGA and DTG curves are summarized in Table 3. of melting enthalpy of 100% crystalline PTMEG [40]. From Table 2 it is
Theoretical value of mass losses of the composites at Tdp= 408.3 °C clear that GO-PPD has a positive effect on degree of crystallinity of
are calculated by using Eq. (2) and the same at Tr= 650 °C are calcu- GPU. The results observed in DSC with 0.10 wt% GO-PPD was com-
lated by using Eq. (3). Theoretical value of filler content (wt%) in the pared with similar wt% loading of unmodified GO to check the effect of
composites are calculated by using Eq. (4). After comparing the theo- GO modification in thermal properties of GPU. From the result tabu-
retical value of filler wt% with the experimental filler loading in the lated in Table 2 it is concluded that GO had significant influence in
composites, it is clear that there is only around 20% deviation. So, it can enhancing the degree of crystallinity of GPU but GO-PPD has even a
be concluded that the fillers are well dispersed within the polymer better influence than GO. Further, GO-PPD has a better effect in glass
matrix by the current synthesis technique. All the theoretical results are transition temperature of polymer also.
presented in Table 4.

Table 3
Mass drop and mass residue calculation from TGA and DTG curves.
Sample ID GPU-0 GPU/GO-PPD 0.05 GPU/GO-PPD 0.10 GPU/GO-PPD 0.20 GPU/GO 0.10 GO-PPD

Total mass of sample (mg) 10.17 10.28 10.33 10.32 10.29 10.55
Mass drop (mg) at Tdp= 408.3 °C 5.77 5.31 3.87 3.41 4.28 2.36
Mass residue (mg) at Tdp= 408.3 °C 4.40 4.97 6.46 6.91 6.01 8.19
Mass drop (%) at Tdp= 408.3 °C 56.7 51.65 37.46 33.04 41.59 22.37
(Mdp) (Mdf)
Mass residue (mg) at Tr= 650 °C 0.08 1.12 0.610 0.740 0.224 1.90
Mass residue (%) at Tr= 650 °C 0.786 10.89 5.90 7.17 2.17 18.01
(Mrp) (Mrf)

71
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Table 4
Comparison of experimental and theoretical value of filler content in composites.
Sample ID GPU/GO-PPD 0.05 GPU/GO-PPD 0.10 GPU/GO-PPD 0.20 GPU/GO 0.10

Initial composite mass (mg) mc 10.28 10.33 10.32 10.29

Initial filler mass (mg) mf 5.14 × 10−3 10.33 × 10−3 20.64 × 10−3 10.29 × 10−3
Initial polymer mass (mg) mp 10.2748 10.3197 10.2994 10.2797
Theoretical mass loss of composite at Tdp= 408.3 °C (mdcfrom Eq. 2) 5.826 5.85 5.84 5.83
Theoretical mass loss of composite at Tr= 650 °C (mrcfrom Eq. 3) 81.68 × 10−3 82.97 × 10−3 84.67 × 10−3 82.65 × 10−3
Theoretical filler mass content in composite (%) (Pffrom Eq. 4) 68.47 × 10−3 120.57 × 10−3 221.32 × 10−3 118.64 × 10−3
Experimental Filler mass content in composite (%) 0.05 0.10 0.20 0.10
Deviation error (%) 36.94 20.57 10.66 18.64

4.7. Dynamic mechanical thermal analysis (DMTA) reason is the saturation of epoxy rings of GPU at.
higher loading. In order to justify the necessity of PPD modification,
Dynamic mechanical thermal analysis (DMTA) was used to de- the results obtained at 0.10 wt% loading of GO-PPD was compared with
termine the viscoelastic properties of the synthesized materials. DMTA that obtained with 0.10 wt% loading of GO. From DMTA results it is
detects different types of transitions and relaxations relating to the clear that GO-PPD provide better reinforcement to GPU compared to
structure and morphology of the samples. Variation of storage modulus GO at similar loading. This was also verified earlier from the tensile
(E′) and damping factor (tan δ) with temperature are represented in properties. Although, the value of storage modulus of GPU/GO 0.10
Fig. 15 and the results are summarized in Table 5. From the results, it is was higher than the GPU/GO-PPD 0.05 and GPU/GO-PPD 0.20. This
clear that storage modulus of GPU/GO-PPD nanocomposites increases should be the H-bonding interactions between the urethane linkage of
with increase in filler loading. This was due to increased interaction GPU and the oxygen containing functional groups of GO which was
between GPU and GO-PPD via physical and chemical bonding [11,18]. lower in number due to partial reduction of GO during functionaliza-
After 0.10 wt% loading of GO-PPD, the E′ value decreases. This was tion [20]. Temperature has a significant effect on the storage modulus
probably due to increased filler-filler interaction compared to filler- value. From Table 5 it is quite clear that storage modulus increases with
matrix interaction at higher loading which leads to filler aggregation the decrease in temperature. This was due to the limitation of the
and hence a reduction in mechanical properties [4]. Another possible molecular chain movement at lower temperature [4]. Damping factor

Fig. 14. (a) DSC first heating curve to show the glass transition temperature (Tg); (b) cooling curve for crystallization and (b) second heating curve for melting of GPU
and its nanocomposites.

72
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Fig. 15. DMTA curves for GPU and its nanocomposites: (a) logarithm of storage modulus (E′) vs. temperature and (b) tan δ vs. temperature.

Table 5
DMTA results of GPU and its nanocomposites.
Samples Tg (°C) Log E′ (MPa) at −75 °C Log E′ (MPa) at −25 °C Log E′ (MPa) at +25 °C

GPU-0 −61.8 3.31 2.28 1.06

GPU/GO-PPD 0.05 −60.4 3.56 2.40 1.12
GPU/GO-PPD 0.10 −58.2 3.80 2.47 1.31
GPU/GO-PPD 0.20 −59.2 3.65 2.50 1.18
GPU/GO 0.10 −60.6 3.70 2.56 1.24

or tanδ represents the ratio of viscous to elastic response of a viscoe- Conflict of interest
lastic material and the peak position is called glass transition tem-
perature (Tg). With increase in GO-PPD loading, Tg of soft segment of The authors declare no conflict of interests.
GPU increases and the damping capacity or the height of tanδ de-
creases, indicating restricted motion of polymer chains [4,7]. This is Acknowledgments
due to stronger chemical bonding between the NH2 group of GO-PPD
and the epoxy group of GPU as well as the H-bonding interaction be- The fanatical support to Dr. Madhab Bera for his fellowship from
tween the urethane linkage of GPU and oxygen containing functional Ministry of Human Resource Development (MHRD), India is gratefully
groups of GO-PPD. Glass transition temperature obtained from DMTA acknowledged.
was little lower than that of DSC. This was due to the difference in
principles of the two testing methods [33]. The cross linking reaction Appendix A. Supplementary data
exerted by HMD was also responsible for this enhancement in modulus
and reduction in damping capacity of nanocomposite. Supplementary data to this article can be found online at https://
This type of results was observed in our previous work also for GO/ [Link]/10.1016/[Link].2019.03.008.
RGO filler thermoplastic polyurethane nanocomposites [4].
References

5. Conclusions [1] G. Oertel, Polyurethane Handbook, 2nd edn, Carl Hanser Verlag, Munich, 1994.
[2] C. Hepburn, Polyurethane Elastomers, Springer Science & Business Media, 2012.
[3] M. Bera, P.K. Maji, Graphene-based polymer nanocomposites: materials for future
In this study, the thermal and mechanical property of in-situ syn- revolution, MOJ Poly Sci. 1 (3) (2017) 94–97, [Link]
thesized glycidol terminated polyurethane and PPD functionalized GO 2017.01.00013.
(GO-PPD) has been discussed. The results obtained with GO-PPD filled [4] M. Bera, P.K. Maji, Effect of structural disparity of graphene-based materials on
thermo-mechanical and surface properties of thermoplastic polyurethane nano-
nanocomposite was also compared with that of GO filled nanocompo- composites, Polymer 119 (2017) 118–133.
sites; to check the effect of PPD modification. It was found that PPD [5] P.K. Maji, P.K. Guchhait, A.K. Bhowmick, Effect of nanoclays on physico-mechan-
modification has a positive impact both in term of thermal and me- ical properties and adhesion of polyester-based polyurethane nanocomposites:
structure–property correlations, J. Mater. Sci. 44 (21) (2009) 5861–5871.
chanical properties of GPU. Mechanical properties were found to in- [6] A.M. Hezma, I.S. Elashmawi, E.M. Abdelrazek, A. Rajeh, M. Kamal, Enhancement of
crease remarkably with the addition of just 0.1 wt% GO-PPD. Tensile the thermal and mechanical properties of polyurethane/polyvinyl chloride blend by
strength increases by 123% and Young's modulus increases by 158% at loading single walled carbon nanotubes, Prog. Nat. Sci. 27 (3) (2017) 338–343.
[7] P. Pokharel, S. Choi, The effect of hard segment length on the thermal and me-
0.10 wt% loading of GO-PPD. Exceptional increase in mechanical
chanical properties of polyurethane/graphene oxide nanocomposites, Compos. A:
properties was due to the synergistic effect of partial cross-linking by Appl. Sci. Manuf. 69 (2015) 168–177.
HMD as well as physical and chemical interactions between GPU and [8] M. Strankowski, P. Korzeniewski, J. Strankowska, A. AS, S. Thomas, Morphology,
GO-PPD. Thermal stability of the resultant nanocomposite was also mechanical and thermal properties of thermoplastic polyurethane containing re-
duced graphene oxide and graphene nanoplatelets, Materials 11 (1) (2018) 82.
found to increase significantly. Temperature for maximum degradation [9] X. Chen, J. Wang, J. Zou, X. Wu, X. Chen, F. Xue, Mechanical and thermal prop-
increases around 16 °C with the addition of 0.2 wt% GO-PPD. The high- erties of functionalized multiwalled carbon nanotubes and multiwalled carbon
performance engineering material has great application opportunity in nanotube–polyurethane composites, J. Appl. Polym. Sci. 114 (6) (2009)
3407–3413.
paints, anticorrosive coatings, and others. [10] G. Mondragon, A. Santamaria-Echart, M.E.V. Hormaiztegui, A. Arbelaiz, C. Peña-

73
M. Bera, et al. Reactive and Functional Polymers 139 (2019) 60–74

Rodriguez, V. Mucci, M. Corcuera, M.I. Aranguren, A. Eceiza, Nanocomposites of [26] J. Liu, H. Yang, S.G. Zhen, C.K. Poh, A. Chaurasia, J. Luo, X. Wu, E.K.L. Yeow,
waterborne polyurethane reinforced with cellulose nanocrystals from sisal fibres, J. N.G. Sahoo, J. Lin, Z. Shen, A green approach to the synthesis of high-quality
Polym. Environ. 26 (5) (2018) 1869–1880. graphene oxide flakes via electrochemical exfoliation of pencil core, RSC Adv. 3
[11] S. Thakur, N. Karak, Ultratough, ductile, castor oil-based, hyperbranched, poly- (29) (2013) 11745–11750.
urethane nanocomposite using functionalized reduced graphene oxide, ACS [27] T. Ramanathan, A.A. Abdala, S. Stankovich, D.A. Dikin, M. Herrera-Alonso,
Sustain. Chem. Eng. 2 (5) (2014) 1195–1202. R.D. Piner, D.H. Adamson, H.C. Schniepp, X. Chen, R.S. Ruoff, I.A. Aksay,
[12] P. Pokharel, High performance polyurethane nanocomposite films prepared from a R.K. Prud'Homme, L.C. Brinson, S.T. Nguyen, Functionalized graphene sheets for
masterbatch of graphene oxide in polyether polyol, Chem. Eng. J. 253 (2014) polymer nanocomposites, Nat. Nanotechnol. 3 (6) (2008) 327.
356–365. [28] L. Stobinski, B. Lesiak, A. Malolepszy, M. Mazurkiewicz, B. Mierzwa, J. Zemek,
[13] K.K. Sadasivuni, D. Ponnamma, B. Kumar, M. Strankowski, R. Cardinaels, P. Jiricek, I. Bieloshapka, Graphene oxide and reduced graphene oxide studied by
P. Moldenaers, Y. Grohens, Dielectric properties of modified graphene oxide filled the XRD, TEM and electron spectroscopy methods, J. Electron Spectrosc. Relat.
polyurethane nanocomposites and its correlation with rheology, Compos. Sci. Phenom. 195 (2014) 145–154.
Technol. 104 (2014) 18–25. [29] N.R. Wilson, P.A. Pandey, R. Beanland, R.J. Young, I.A. Kinloch, L. Gong, Z. Liu,
[14] X. Wang, Y. Hu, L. Song, H. Yang, W. Xing, H. Lu, In situ polymerization of gra- K. Suenaga, J.P. Rourke, S.J. York, J. Sloan, Graphene oxide: structural analysis and
phene nanosheets and polyurethane with enhanced mechanical and thermal prop- application as a highly transparent support for electron microscopy, ACS Nano 3 (9)
erties, J. Mater. Chem. 21 (12) (2011) 4222–4227. (2009) 2547–2556.
[15] H. Kalita, N. Karak, Epoxy modified bio-based hyperbranched polyurethane ther- [30] S. Zhang, N. Pan, Supercapacitor performance of crumpled and planar graphene
mosets, Des. Monomers Polym. 16 (5) (2013) 447–455. materials produced by hydrogen gas reduction of graphene oxide, J. Mater. Chem.
[16] H.C. Hsia, C.C.M. Ma, M.S. Li, Y.S. Li, D.S. Chen, Glycidyl-terminated polyurethane A 1 (27) (2013) 7957–7962.
modified epoxy resins: mechanical properties, adhesion properties, and mor- [31] T. Mondal, S. Basak, A.K. Bhowmick, Ionic liquid modification of graphene oxide
phology, J. Appl. Polym. Sci. 52 (8) (1994) 1137–1151. and its role towards controlling the porosity, and mechanical robustness of poly-
[17] H. Yeganeh, M.M. Lakouraj, S. Jamshidi, Synthesis and properties of biodegradable urethane foam, Polymer 127 (2017) 106–118.
elastomeric epoxy modified polyurethanes based on poly (ε-caprolactone) and poly [32] P.K. Maji, P.K. Guchhait, A.K. Bhowmick, Effect of the microstructure of a hyper-
(ethylene glycol), Eur. Polym. J. 41 (10) (2005) 2370–2379. branched polymer and nanoclay loading on the morphology and properties of novel
[18] A. Hussein, S. Sarkar, D. Oh, K. Lee, B. Kim, Epoxy/p-phenylenediamine functio- polyurethane nanocomposites, ACS Appl. Mater. Interfaces 1 (2) (2008) 289–300.
nalized graphene oxide composites and evaluation of their fracture toughness and [33] J. Zhang, C. Zhang, S.A. Madbouly, In situ polymerization of bio-based thermo-
tensile properties, J. Appl. Polym. Sci. 133 (34) (2016). setting polyurethane/graphene oxide nanocomposites, J. Appl. Polym. Sci. 132 (13)
[19] M. Bera, P. Gupta, C. Yadav, P.K. Maji, Facile one-pot synthesis of graphene oxide (2015).
by sonication assisted Mechanochemical approach and its surface chemistry, J. [34] P. Bandyopadhyay, W.B. Park, R.K. Layek, M.E. Uddin, N.H. Kim, H.G. Kim,
Nanosci. Nanotechnol. 18 (2) (2018) 902–912. J.H. Lee, Hexylamine functionalized reduced graphene oxide/polyurethane nano-
[20] H.L. Ma, H.B. Zhang, Q.H. Hu, W.J. Li, Z.G. Jiang, Z.Z. Yu, A. Dasari, composite-coated nylon for enhanced hydrogen gas barrier film, J. Membr. Sci. 500
Functionalization and reduction of graphene oxide with p-phenylenediamine for (2016) 106–114.
electrically conductive and thermally stable polystyrene composites, ACS Appl. [35] S. Wu, T. Shi, L. Zhang, Preparation and properties of amine-functionalized reduced
Mater. Interfaces 4 (4) (2012) 1948–1953. graphene oxide/waterborne polyurethane nanocomposites, High Perform. Poly. 28
[21] M. Kotal, S.S. Banerjee, A.K. Bhowmick, Functionalized graphene with polymer as (4) (2016) 453–465.
unique strategy in tailoring the properties of bromobutyl rubber nanocomposites, [36] A.V. Raghu, Y.R. Lee, H.M. Jeong, C.M. Shin, Preparation and physical properties of
Polymer 82 (2016) 121–132. waterborne polyurethane/functionalized graphene sheet nanocomposites,
[22] R. Zhong, Z. Zhang, H. Zhao, X. He, X. Wang, R. Zhang, Improving thermo-oxidative Macromol. Chem. Phys. 209 (24) (2008) 2487–2493.
stability of nitrile rubber composites by functional graphene oxide, Materials (Basel, [37] Z. Chen, H. Lu, Constructing sacrificial bonds and hidden lengths for ductile gra-
Switzerland) 11 (6) (2018). phene/polyurethane elastomers with improved strength and toughness, J. Mater.
[23] D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, Chem. 22 (25) (2012) 12479–12490.
L.B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano 4 [38] X. Wang, D. Tan, Z. Chu, L. Chen, X. Chen, J. Zhao, G. Chen, Mechanical properties
(8) (2010) 4806–4814. of polymer composites reinforced by functionalized graphene prepared via direct
[24] B. Yuan, C. Bao, L. Song, N. Hong, K.M. Liew, Y. Hu, Preparation of functionalized exfoliation of graphite flakes in styrene, RSC Adv. 6 (113) (2016) 112486–112492.
graphene oxide/polypropylene nanocomposite with significantly improved thermal [39] O. Nabinejad, D. Sujan, M.E. Rahman, I.J. Davies, Determination of filler content
stability and studies on the crystallization behavior and mechanical properties, for natural filler polymer composite by thermogravimetric analysis, J. Therm. Anal.
Chem. Eng. J. 237 (2014) 411–420. Calorim. 122 (1) (2015) 227–233.
[25] S.K. Singh, M.K. Singh, P.P. Kulkarni, V.K. Sonkar, J.J. Grácio, D. Dash, Amine- [40] K.N. Raftopoulos, B. Janowski, L. Apekis, K. Pielichowski, P. Pissis, Molecular
modified graphene: thrombo-protective safer alternative to graphene oxide for mobility and crystallinity in polytetramethylene ether glycol in the bulk and as soft
biomedical applications, ACS Nano 6 (3) (2012) 2731–2740. component in polyurethanes, Eur. Polym. J. 47 (11) (2011) 2120–2133.

74

View publication stats