Polymer 100 (2016) 194e205

Contents lists available at ScienceDirect

Polymer
journal homepage: www.elsevier.com/locate/polymer

Molecular dynamics simulations of polyvinyl acetate-perfluorooctane

based anti-stain coatings
Nityanshu Kumar, Gaurav Manik*
Department of Polymer and Process Engineering, Indian Institute of Technology Roorkee, Saharanpur Campus, Saharanpur 247001, India

a r t i c l e i n f o a b s t r a c t

Article history: Polyvinyl acetate (PVAc)-Perfluorooctane (PFO) systems are proposed as potential future anti-stain and
Received 28 June 2016 easy-to-clean coating materials. Such coatings possess film forming (due to the PVAc content) and anti-
Received in revised form stain property (due to the presence of PFO) could find potential applications as useful coatings for res-
5 August 2016
idential and industrial use. In this work, we present a simulation strategy to generate equilibrated
Accepted 6 August 2016
structures of the proposed coatings of different PFO content in PVAc. Important coatings properties, such
Available online 11 August 2016
as surface free energy, energy of interaction with a metal substrate or substrate wetting, contact angles of
oil and water, are estimated from simulations and compared with the properties of equivalent systems in
Keywords:
Interaction energy
bulk state. The density, solubility parameter, surface energy of the simulated systems and water and oil
Molecular dynamics contact angles, compare very well with the available literature estimates. It is observed that with
Oleophobic coating increasing PFO content, the surface energy of the simulated coating surfaces significantly reduces due to
Perfluoropolymers the surface aggregation of fluorine based entities, and this reflects directly in a visible and desirable
Perfluorooctane increase in contact angles of oil and water. In contrast to this, it is observed that the interaction energy of
Surface aggregation the structures against a simulated aluminum substrate undesirably decreases with increasing PFO con-
Surface free energy tent, and hence, works oppositely to the surface energy variation. Further, the interaction energy changes
significantly beyond the 45.8 wt% PFO content. These results help us to screen and propose a potential
useful coating formulation of ~35.2 wt% PFO in PVAc, well optimized for strong oleophobicity and
hydrophobicity.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction has been shown to possess many desirable properties by Reisinger

et al. [4]. The high extent of electronegativity and low polarizability
The synthesis and characterization of anti-stain coatings has of fluorine atom induces hydrophobic and oleophobic properties to
attracted the interest of scientists worldwide since a long time. The the coatings synthesized from them. Further, Prathab et al. [5] have
stain, in most cases, arises on a surface due to the deposition of shown that the surface aggregation reduces the surface energy
molecules that are of oil-like nature. Due to this pertinent reason, which decreases linearly with increasing fluorine content in the
the importance of oleophobic coatings has grown in the recent synthesized polymer blends or copolymers. The importance of
times, and a lot of effort has been made in the recent past to un- fluoropolymers to exhibit a great deal of oleophobicity has been
derstand the fundamental reasons behind why these coatings extensively investigated in the recent past. For example, Lakshmi
behave the way they do. et al. [6] found that sol-gel nanocomposite coatings made with
There has been an immense interest in the development of anti- perfluoroalkyl methacrylic copolymer as an additive increased the
stain polymeric coatings in the past [1,2]. Interestingly, many flu- lubricant oil contact angle from <10
to >113
. They have attributed
oropolymers, especially perfluoropolymers, have been found to this to the lowering of surface energy due to the addition of the
reduce the surface wettability towards water and oil [3]. The fluoropolymer and silica aggregates. Fabbri et al. [7] synthesized
presence of atomic fluorine in the backbone or in the side chains perfluoropolyether (PFPE) based organic-inorganic hybrid coatings
for glass substrates using sol-gel process. They observed that the
surface segregation of perfluoro based PFPE segments was highly
* Corresponding author. increased with the use of PFPE in inorganic matrix, primarily when
E-mail addresses: kumar.nityanshu@gmail.com (N. Kumar), gauravmanik3m@ manually coated on a glass substrate, leading to high values of
gmail.com, manikfpt@iitr.ac.in (G. Manik).

http://dx.doi.org/10.1016/j.polymer.2016.08.019
0032-3861/© 2016 Elsevier Ltd. All rights reserved.
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 195

contact angles against n-hexadecane and water, thereby exhibiting, recognized. The simulations can help to tailor the underlying
oleophobic/lipophobic and hydrophobic behavior respectively. polymer formulations to meet the application requirements
Cansoy et al. [8] studied the contact angles of copolymer surfaces without actually spending huge human efforts in synthesizing
against low surface energy (LSE) liquids of different chain lengths them. Molecular dynamics (MD) is one such simulation technique
such as octane, decane, tetradecane and hexadecane. The oleo- that works on the virtual simulation of physical arrangements of
phobicity of the surfaces was found to be dependent on the liquid atoms and molecules. The atoms, and hence, molecules or polymers
surface tension and wt% of Zonyl-TM used. They found that the made from them are permitted to intermingle for a period of time,
liquid chain length controls the cohesion among its molecules, and with their motion governed through well-defined Newton's equa-
hence, a larger chain length liquid would be expected to have high tions of motion. The procedure is repeated for a large number of
cohesion, and hence, increased contact angles when applied on a steps, with the iterative calculations of bonded and non-bonded
surface. forces, till a more realistic material structure is generated. MD has
Apart from the increasing worldwide interest to formulate been increasingly used to determine useful polymer properties of
coatings that provide oil repellency, several attempts have been interest in practical applications [14]. Several researchers have used
made in the past to synthesize superhydrophobic coatings as well, molecular mechanics (MM) and MD simulations in the past to
that would be used to impart anti-corrosion and anti-biofouling predict coatings or thin films behavior at molecular level. For
properties to surfaces. The concepts of hydrophilic and hydropho- instance, Prathab and co-workers [3] have molecularly simulated
bic capacity apply not only to bodies and their surfaces but also to and analyzed equilibrated amorphous structures of perfluoroalkyl
molecules. Thus, in molecules of surface-active substances, a methaacrylates and copolymers derived from methyl methacrylate
distinction is made between hydrophilic (polar) and hydrophobic (MMA) and 1,1-dihydroperfluorohendecyl methacrylate (F10MA).
(hydrocarbon) groups. The hydrophilic capacity of the body surface They found that the surface energy of the structures decreased as
can change markedly through the absorption of such substances. the number of perfluoroalkyl groups in the copolymers increased.
An increase in the hydrophilic capacity is called hydrophilization, They further observed that the dominant contributions in surface
and a decrease is called hydrophobization. Both the phenomena interactions are van der Waals and Coulombic energies. Recently,
play an important part in the beneficiation of ores by flotation. In Dalvi et al. [15] have used atomistic MD to explain the underlying
textile technology, the hydrophilization of cloth (fibers) is neces- reasons of hydrophobicity and oleophobicity. They explain that the
sary for successful dyeing, bleaching, and laundering, and hydro- hydrophobicity of the fluoropolymers is due to their inherent
phobization is necessary to impart water and moisture resistance to “fatness”, i.e., the tendency to pack with relative less density at the
cloth. surface than the corresponding hydrocarbons. Likewise, other
Hydrophobic coatings, as the name suggests are water- noteworthy attempts in using molecular simulations are for esti-
repellent, with the water molecules attracted to the coatings, mation of compatibility of polymer blends. For example, Sheetal
extremely weakly. These are produced in the form of mono- et al. have attempted to predict the compatibility of poly(vinyl
molecular layers or lacquer films by treating a material with solu- alcohol) and chitosan blends [16], of poly(L-lactide) and poly(vinyl
tions, emulsions, or less frequently, vapors of hydrophobic agents, alcohol) blends [17] and of poly(vinyl alcohol) and poly(methyl
which are substances that intermingle weakly with water [9]. metaacrylate) blends [18]. Further, molecular simulations have
Substances used as hydrophobic agents often include salts of fatty enabled researchers to estimate surface properties such as surface
acids and such metals as copper, aluminum, and zirconium; cation- energy of blends [19] and gas diffusion properties of films [20].
active surface-active agents; and low- and high-molecular-weight In this work, we attempt to molecularly simulate equilibrated
organosilicon and organic fluorine compounds. Hydrophobic structures of coatings of different perfluorooctane (PFO) content in
coatings protect several materials (metal, wood, plastics, leather, polyvinyl acetate (PVAc) and to determine their physical and sur-
and fabric and non-fabric fibrous materials) from the destructive face properties. The PVAc provides the film forming property (i.e.
act of water or wetting [10]. Polizos et al. [11] synthesized super- good adhesion or interaction energies with the metal substrate)
hydrophobic coatings from diatomaceous earth (DE). The nano- and its emulsions and colloidal dispersion coatings are well-known
structured spherical and cylindrical structures in DE particles were for high flexibility to withstand mechanical stress, high gloss, anti-
fluorinated to create coatings with superhydrophobic properties. shrink properties. Kollicoat® SR 30 D, an aqueous polyvinyl acetate
The abrasion resistance and longtime superhydrophobic property dispersion is an interesting new dispersion because of its physi-
was proposed to be dependent on the surface morphology of the cochemical properties and its flexible extended release properties
nanostructure silica particles. [21]. The choice of PFO as an additive is to render the anti-stain
Recent literature shows that there is an increased interest in property to the coating for potential applications in residential
synthesizing and characterizing coatings that possess both super- and industrial use. We have made a hypothesis by considering a
hydrophobicity and superoleophobicity. Nugyena et al. [12], for pure PFO system as a coating surface to observe any saturation
example, have created superhydrophobic (and superoleophobic) trend with increasing PFO content. Further, although there is an
coatings using electroless etching of nanostructured silicon sur- increased interest in the development of C4-C6 chemistry for fluoro
faces. They found that depositing silver particles on the etched or allied polymers for anti-stain coatings due to environmental
surfaces could be used to increase the surface roughness, and issues, PFO is specifically chosen in this study as it does not possess
hence, attain reduced surface wettability and increased contact such disadvantages and has also been accredited to be a breathable
angles of water and low surface energy liquids such as oil. Cenzig fluid in partial liquid ventilation [22]. PFO being a small molecule as
et al. [13] argue that the increase in number of -(CF2)- groups in the compared to complex fluoropolymers can easily migrate to the
side chain contributes to a decrease in surface free energy of the surface, can also be used with the polymers of high Tg due to its
resultant polymer surfaces, thereby, increasing water and oil high mobility and possesses a good service life if compatible with
repellency properties. the system in which it is incorporated. This work employs molec-
While a lot of mentioned efforts have been made to synthesize ular simulations to predict the surface properties of proposed
and test such useful coatings, the importance of molecular simu- coating namely contact angle, surface energy and adhesion energy
lations as a potential tool in understanding the surface morpho- with the substrate that may serve as a useful study to researchers to
logical properties of thin films or coatings has been duly optimize similar coating formulations in future.
196 N. Kumar, G. Manik / Polymer 100 (2016) 194e205

2. Simulation methodology minimized separately using MM. The energy minimized structures
of PVAc (10 chains) and PFO (in different wt%) are used for the
Molecular simulations have been performed using commercially construction of different amorphous cells. Amorphous Cell module
available software, Material Studio 7.0, purchased from Accelrys Inc. of Material Studio has been used to construct initial realistic 3D
[23]. Interactions among the various atoms are calculated using periodic structures. The module builds molecules in a cell in a
COMPASS (condensed-phase optimized molecular potentials for Monte Carlo fashion, by minimizing close contacts between atoms
atomistic simulation studies) force-field [9,24]. COMPASS enables and ensuring a realistic distribution of torsion angles for the given
accurate and simultaneous prediction of gas-phase properties force-field. To avoid limited cell size effects, periodic boundary
(structural, conformational, vibrational, etc.) and condensed-phase conditions (PBC) in which the central cell repeats itself in three
properties (equation of state, cohesive energy, etc.) for a broad dimensions (x, y, z), has been incorporated.
range of molecules. The energy of the structures is minimized using The initially generated high energy periodic amorphous cells
a “Smart Minimizer” function of “Discover” module in Material (with different PFO wt%) are subjected to energy relaxation using
Studio that uses a set of three methods- Steepest Descent, Conjugate the Smart Minimizer module for 5000 MM iterations. The energy
Gradient and Newton, to minimize the energy of the initial minimized structures are then subjected to equilibration by NVT
randomly generated structure. The temperature in all simulations is (constant number of particles, volume and temperature) MD sim-
maintained constant with the Andersen thermostat and pressure ulations, to attain structures with realistic density and low
kept controlled in NPT (constant number of particles, pressure and potential-energy. For this, a high temperature (750 K) faster NVT
temperature) MD simulations using Berendsen algorithm. The ve- dynamics was run for about 50 ps to make the system forget the
locity Verlet algorithm is used to integrate the equations of motion. initial configuration bias. After this, NVT simulations are performed
Non-bonded interactions have been calculated using the group- at 298 K for a period of 300 ps with a time step of 1 fs, and the
based approach with a cutoff distance of 9.50 Å. The group based generated frames saved after every 0.1 ps. From the set of 3000
approach is considered to be the best approach for the neutral frames generated, the lowest potential energy frame is selected and
fragments with dipoles as found earlier as well [3]. The Amorphous further minimized using MM to a convergence of 0.01 kcal/mol/Å.
Cell and Forcite modules of Material Studio have been used in this The generated frame from MM is subjected to NPT dynamics for
work to create the simulation cells and run atomistic MD. Fig. 1 300 ps at 298 K and 0.0001 GPa (1 atm) to eliminate the local
shows the flow chart of simulation protocol used for estimating system stresses and equilibrate the density.
the relevant coating properties. The amorphous cells obtained from the procedure are equili-
brated cells and have been used for all further property estimations.
2.1. Generation of equilibrated bulk structures The edge lengths of the finally generated central amorphous cells
ranged from 20.002 to 21.533 Å for 1000% PFO in PVAc matrix. An
Chains of PVAc (Chain length, n ¼ 16) with isotactic stereo- amorphous cell model of the generated PVAc-PFO structure with
chemical structure and molecules of PFO are generated and energy 24.1% PFO, as an example, has been shown in Fig. 2.

Fig. 1. Flow chart of the simulation strategy to estimate characteristic properties of PVAc-PFO based coatings.
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 197

NPT MD simulations are further performed on the equilibrated minima. A high temperature is chosen to shake the cells well and to
cells at 298 K and 0.0001 GPa (1 atm) for 30 ps to equilibrate the remove the packing inefficiencies that could have occurred due to
density. The frames are stored after every 0.1 ps. The average of the the elongation process. This step is followed by a 300 ps MD
densities of all the saved equilibrated frames for all the different simulation run at 298 K. Frames after every 0.01 ps are captured
PVAc-PFO systems, are reported in Table 1 with the corresponding and the lowest energy snapshot minimized with the convergence
standard deviations. The deviations in the reported densities are order of 0.01 kcal/mol/Å to get the finally energy minimized thin
found to lie within 1.3% of average values indicating a good density film based conformations. The surface energy, g, of the generated
equilibration. The simulated densities of the equilibrated structures films is then calculated using the equation given below:
are compared with the experimental values for PVAc [25] and PFO  
reported previously [10], and also with values computed using the EThinfilm  EAmorphous cell
linear density equation reported by Brice and Coon [26]. The g¼ (1)
2A
simulated densities are found to be in close vicinity to the experi-
mental ones, thereby, validating the simulation protocol. The den- where, EThin film represents the energy of the film, and EAmorphous cell
sities of the hybrid PVAc-PFO systems are not available in literature, the energy of equilibrated bulk amorphous cells, and 2A is the
and hence, could not be compared in this work. surface area of the two surfaces each of area A formed due to cre-
ation of film.

2.2. Generation of thin films (free surfaces)

2.3. Simulation of coating-decane and coating-water systems
The material surface properties of thin films or coatings are
important primarily when considering applications in the fields of Dirt-repellency is one of the most important properties of the
adhesion, coating, paints and microelectronics. The films are coatings. Since the oil presence on a surface increases its chances of
considered to be having two surfaces and experience a directional attracting dirt, an estimation of its oleophobic property is impor-
asymmetry during interaction with their surroundings. With a view tant to capture surface dirt-repellency. Likewise, a hydrophobic
that MM and MD simulations can give vital surface morphological surface is known to be easy-cleanable when washed or splashed
information of films, amorphous structures of thin films are with water. Hence, there is a need to estimate the hydrophobic
generated. property of the proposed coatings as well.
The generated equilibrated cells mentioned in section 2.1 are In the present work, the hydrophobic and oleophobic nature
elongated along one of the directions (z-axis) till the molecules of have been confirmed by determining the concentration profile of
the simulation cells are found not to interact with their image along water and decane molecules respectively, on the created coating
the elongated direction. The z dimensions of the cells are elongated surface. The choice of decane molecules has been made as it can be
to a reasonable spacing of 100 Å to form thin films of different presumed to exhibit a reasonable oil-like behavior. The Forcite
PVAc-PFO systems. Thin films are constructed according to the module of Material Studio 7.0 has been used for running MM and
methodology used previously [27e29] for other polymeric systems. MD simulations of the two layered systems. The equilibrated cells
After the formation of these thin films from the bulk, a high tem- of PVAc-PFO (with different wt% PFO) form one of the layers, Layer-
perature (750 K) NVT MD for ~ 50 ps, and further, MM energy 1. Separate cells of water and decane molecules are constructed and
minimization eliminated the structures from the local energy equilibrated using the procedure defined in section 2.1, and form
the Layer-2 in the respective two layered systems. For this purpose,
about 1000 molecules of water and 300 molecules of decane have
been considered. Layer-1 and Layer-2 of the different systems are
separated by a vacuum of 5 Å, and a vacuum of 100 Å is introduced
above Layer-2, so that the Layer-2 does not interact with the peri-
odic image of Layer-1.
To compare the hydrophobic and oleophobic nature of the
PVAc-PFO based coating with that of the pure PVAc based coating,
the movement of the cluster of decane and water molecules is
separately simulated. The finally formed two layered systems
mimic the condition of the water on coating (shown in Fig. 3) and
decane on coating (shown in Fig. 4) situations. Fig. 3(a) shows
movement of water molecules against the simulated two layered
structures after 10 ps simulation steps, when it is exposed to a pure
PVAc coating as against when it is exposed to PVAc with 55.9% PFO
(Fig. 3(b)). The figure clearly shows that at similar time steps the
water molecules move faster towards the coating in case of pure
PVAc coating as against PVAc-PFO systems, thereby, indicating the
improved hydrophobicity of the latter. Likewise, Fig. 4(a) and (b)
show the simulated two layered structures which mimic the sur-
faces being exposed to oil (decane) for the case of pure PVAc and
55.9% PFO-PVAc systems after a simulation time of 10 ps. The figure
clearly shows that pure PVAc attracts the decane cluster much
faster than the system that contains 55.9% PFO. This helps us to
conclude the improved oleophobicity of the latter system.
Fig. 2. The amorphous model of PVAc-PFO system with 24.1% PFO (carbon atoms e
grey, hydrogen e white, oxygen e red and fluorine e pale blue color). (For interpre-
To quantify the hydrophobic and oleophobic nature of the
tation of the references to colour in this figure legend, the reader is referred to the web simulated surfaces, it is essential to simulate the final structure, and
version of this article.) hence, the contact angles of water and decane on such systems. For
198 N. Kumar, G. Manik / Polymer 100 (2016) 194e205

Table 1
Comparison of simulated densities with experimental and theoretically calculated densities for the PVAc-PFO system for varying PFO concentrations.

PFO content (wt %) Simulated density (g cm3) Experimental/theoretical density (g cm3)

0.0 1.154 ± 0.009 1.19 [25]

12.4 1.202 ± 0.012 Not Available
24.1 1.257 ± 0.008 Not Available
35.2 1.323 ± 0.017 Not Available
45.8 1.379 ± 0.009 Not Available
55.9 1.447 ± 0.012 Not Available
65.5 1.513 ± 0.018 Not Available
100.0 1.778 ± 0.018 1.77 [10], 1.75 [26]

Fig. 3. Comparison of the structure of (a) pure PVAc and (b) PVAc with 55.9 wt % of PFO as Layer-1 with water molecules (in pink) forming Layer- 2, after an MD run of about 10 ps of
water molecules (C atoms e grey, H e white, H of H2O ewhite þ pink, Oered, O of H2O e red þ pink and F e pale blue color). (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)

Fig. 4. Comparison of the structure of (a) PVAc and (b) PVAc with 55.9 wt % of PFO as Layer-1 with decane (in grey) molecules forming Layer-2 after the MD run of about 10 ps
(carbon atoms e grey, hydrogen e white, oxygen e red and fluorine e pale blue color). (For interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)

this, thin film amorphous cells of PFO-PVAc systems are used to interface. The interaction between the polymeric coating and the
build their respective super cells that have x- and y-dimensions as 4 substrate is primarily through van der Waals forces, specifically,
times that of the respective amorphous cells. This is done to ease when the substrate is metal. It is well-known that metals have high
the visualization and calculation of the contact angle, and the surface energy than non-metals, display improved adhesion of
concentration profiles of decane and water molecules, on the coatings and are frequently used to test efficacy of coatings. Hence,
coating surface. a metal namely aluminum (Al) has been taken as a starting sub-
After the formation of two layers, a constraint on atom co- strate for simulations. Al is the most used metallic material besides
ordinates of Layer-1 (PVAc-PFO system) is applied and a geometry steel and it is very easy to mimic the Al surface compared to steel
optimization of decane and water molecules carried out till an using molecular simulations. Further, the reported surface energy
energy convergence of 0.01 kcal/mol/Å is achieved. A NVT MD of Al is ~1.14 J/m2 which is much smaller than that of other
simulation is next performed for 1000 ps with a time step of 1 fs commonly used metals such as copper (1.83 J/m2), iron (2.48 J/m2)
using a Nose
eHoover thermostat (with Q ratio of 0.01) to maintain and titanium (2.10 J/m2) [33]. Before the start of simulations the
the system at 298 K. The output frame is stored after every 100 interface energies of coatingemetal system are not known and so is
steps and the last 250 frames are used for estimation of contact also the fate of interfacial adhesion of coatings to substrates. Since
angles and concentration profiles. Al possesses relatively less surface energy than other metals, hence
it is hypothesized that if the proposed coating would wet and
2.4. Simulation of coating-metal surface interaction adhere strongly to the Al surface, it may wet the other potentially
useful group of metals as well.
A strong adhesion of coating on the surface of metal, polymer or To construct the substrate structure, a FCC (Face Centered Cubic)
any other material is considered as a desirable property. It is crystal lattice cell of aluminum (Al) (lattice parameters:
because of this reason that in the recent past, lot of computational a ¼ b ¼ c ¼ 4.0495 and a ¼ b ¼ g ¼ 90
) is exported from the
efforts have been made to investigate polymer-polymer, polymer- Material Studio 7.0 database. The exported lattice is then cleaved
non polymer and polymer-metal oxide interface interactions along the plane of lowest surface energy, i.e., (1 1 1) for Al [34]. The
[30e32]. The adhesion or interaction energy is the measure of the surface formed (with dimensions, u ¼ v ¼ 2.8634 Å) is then energy
strength of adhesion holding two materials together at the minimized and structure relaxed using MM. COMPASS force-field is
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 199

used with the Ewald summation and atom based methods for in energy per mole of polymer, Ecohesive, if all of its intermolecular
electrostatic and van der Waals forces estimation. Also, only the top forces are eliminated in a unit molar volume. Thus, if the molar
few layers of surface atoms would interact with the polymer and volume of polymer is Vmolar, then CED is given by Eq. (3):
the rest of the atoms could be considered to be part of the bulk, and
therefore, have little effect. For this, a constraint is applied on the CED ¼ ðEcohesive =Vmolar Þ (3)
bulk atoms to that they are not minimized. After which the mini- The Hildebrand solubility parameter (d) is related to CED by the
mization is done using Forcite module. The surface obtained has equation:
very small surface area, and hence, a super cell of the dimension
nearly same or greater than the polymer amorphous cell was made. d ¼ CED1=2 ¼ ðEcohesive =VMolar Þ1=2 (4)
The surface obtained is converted to a crystal using “Build/Vac-
uum slab” function and setting vacuum thickness to 0 Å. The simulated d values are presented in Table 2 and are
The “Layer builder” function is then used to place the polymer compared with the literature estimates for pure PVAc [35e41] and
over simulated Al surface. A large vacuum above the polymer is PFO [42]. The comparison shows that the simulated values closely
introduced so that the polymer does not interact with the image of match with the reported values, thereby, validating the simulation
Al due to PBC. Before initiating the final dynamics step, the entire protocol and results.
set of surface atoms is constrained. Further to this, the geometry of In the considered system, PFO has been assumed to be used as a
the polymer layer with respect to the metal surface is optimized liquid additive to PVAc matrix. Strictly, an additive may show
within the convergence limit of 0.01 kcal/mol/Å. Finally, NVT dy- compatibility with the base polymer if c is negative [43]. To un-
namics is performed at 298 K for 300 ps. Fig. 5 shows the devel- derstand the liquid PFO additive compatibility in polymer, the c
opment of Layer-1 and Layer-2 system, and situation after 50 ps parameter for the different systems has been calculated using the
simulation. The interaction energy, post to equilibration is then procedure given in Appendix-1 (Table 1). For all the formulated
calculated using Eq. (2): systems c has been found to be negative, thereby, demonstrating
  compatibility of PFO in PVAc for the entire wt% range. Table 2
EInteraction ¼ ETotal  EAl Surface þ EPolymer (2) suggests a difference in the solubility of PFO and PVAc which
would lead to the PFO content migrate to the surface readily, and
thereby, contribute to the surface hydrophobicity and
where, ETotal is the combined energy of the Al surface and the PVAc-
oleophobicity.
PFO system, EAl Surface is the energy of the solo Al surface, and
EPolymer is the energy of the PVAc-PFO system taken independently.
3.2. Estimation of surface energy and its components
3. Results and discussion
The surface energy, g, estimated using Eq. (1) (section 2.2) for
3.1. Estimation of solubility parameter different PVAc-PFO systems is reported in Table 3. The simulations
provide a value of 37.2 mJ/m2 for PVAc which is close to the
The solubility parameters have been calculated after estimating experimental value of 36.5 mJ/m2 reported earlier [44]. Likewise,
the cohesive energy density (CED) values. CED refers to the increase the simulated value for pure PFO is 14.19 mJ/m2 which is also close
to the experimental value of 14.5 mJ/m2 reported earlier [45]. The
close match of the simulated values with the experimental ones
validates the efficiency of molecular simulation strategy in esti-
mating the surface energy of PVAc-PFO systems.
Alternatively, surface energy can be calculated from the solu-
bility parameter values using the empirical equation of Zisman [46]
described by Eq. (5).

g ¼ 0:75ðEcohesive Þ2=3 (5)

The MD simulated surface energies are also compared with the
Zisman estimates in Table 3. The simulated surface energy from
molecular simulations for pure PVAc (37.2 mJ/m2) closely matches
with the computed Zisman value of 38.04 mJ/m2 for pure PVAc.
However, with increasing PFO content in PVAc, the MD simulated
values are found to deviate increasingly from Zisman estimates. The
method according to Zisman is scientifically applied for investi-
gating the wettability of solid by determining the critical surface
tension with the help of the contact angle of a liquid drop (pref-
erably non-polar). Since the mentioned simulations estimate the
energy values from quite a different procedure, a reasonable vari-
ation in estimates may be expected.
Fig. 6(a) shows that the variation of surface energy of PVAc-PFO
system follows a decreasing trend with increasing PFO content. The
two unfilled triangles denote the experimental values and the black
dots represent the values founded via simulations. A similar result
Fig. 5. Illustration of (a) The two layered system consisting of aluminum (Al) crystal as was seen experimentally previously [47,48] with fluorinated seg-
the first layer (shown in pink mini-spheres) and PVAc containing 55.9% PFO at time
t ¼ 0 ps (b) Final two layered structure post-MD run of 300 ps. (For interpretation of
ments of poly(caprolactone-b-perfluoropolyether- b-caprolactone)
the references to colour in this figure legend, the reader is referred to the web version (PCLePFPEePCL) triblock copolymers aggregating on the surface.
of this article.) Fig. 6(b) shows that the van der Waals and Coulombic electrostatic
200 N. Kumar, G. Manik / Polymer 100 (2016) 194e205

Table 2
Comparison of simulated verses reported solubility parameters for the various PVAc-PFO systems.

PFO (wt%) PFO:PVAc (chain ratio) Cell dimensions (Å) Surface area (m2) CED (J/cm3) Solubility parameter (d) (J/cm3)1/2

Simulated Reported

0 0\10 21.53 463.68 361.19 19.005 ± 0.024 18.61 [34], 19.33 [35],
19.55 [36], 18.96 [37],
19.1e21.6 [38e40]
12.4 2\9 21.40 457.91 338.74 18.405 ± 0.024 Not Available
24.1 4\8 21.31 454.21 319.55 17.876 ± 0.018 Not Available
35.2 6\7 21.11 445.79 304.71 17.456 ± 0.030 Not Available
45.8 8\6 20.95 438.95 284.06 16.854 ± 0.018 Not Available
55.9 10\5 20.84 434.29 273.51 16.538 ± 0.021 Not Available
65.5 12\4 20.73 429.83 251.83 15.869 ± 0.025 Not Available
74.7 14\3 20.56 422.61 228.80 15.126 ± 0.024 Not Available
83.5 16\2 20.53 421.52 206.24 14.361 ± 0.024 Not Available
91.9 18\1 20.16 406.24 202.95 14.246 ± 0.024 Not Available
100 20\0 20.00 400.07 177.72 13.331 ± 0.020 12.9 [42]

Table 3
Comparison of the simulated surface energy (g) and the reported (experimental/estimated) values for different systems with different PFO content in PVAc.

PFO (wt%) PFO:PVAc (chain ratio) Dimensions (Å) Surface energy (mJ/m2)

Simulateda Experimentalb Zismanc estimate [46]

0 0\10 21.53 37.20 36.5 38.04

12.4 2\9 21.40 27.84 e 36.45
24.1 4\8 21.31 24.95 e 35.06
35.2 6\7 21.11 23.40 e 33.96
45.8 8\6 20.95 20.37 e 32.41
55.9 10\5 20.84 19.18 e 31.60
65.5 12\4 20.73 17.50 e 29.91
74.7 14\3 20.56 16.34 e 28.06
83.5 16\2 20.53 15.46 e 26.18
91.9 18\1 20.16 14.52 e 25.90
100 20\0 20.00 14.20 14.5 23.71
a
Estimated from Eq. (1).
b
References [44,45].
c
Estimated from Eq. (5).

Fig. 6. Variation of (a) surface energy, in mJ/m2, as a function of weight % of PFO in PVAc (b) van der Waals and Coulombic energy components, in mJ/m2, with changing PFO
content.

energy terms of surface energy contribute considerably to the for- energy with increasing fluorination, as reflected in Table 3, is in
mation of free surface of the films. Also, with increasing fluorine accordance with the experimental observation previously observed
content, there is a noticeable decrease in van der Waals energy with [49e51].
a systematic increase in Coulombic energy. The overall effect of the
two opposite trends is a decrease in surface energy with increasing
3.3. Evaluation of fluorine profile in bulk and coating
fluorine composition. This is due to a higher weightage of van der
Waals component in overall surface energy, and also, a substantial
Fig. 7 illustrates the profile of relative concentration, R, of fluo-
higher decrease in the van der Waals energy than the increase in
rine atoms within the bulk/thin film measured along the direction
Coulombic energy component. This trend of decrease in surface
perpendicular to the simulated coating structures (z-axis) as a
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 201

function of increasing distance from the center of the thin film/ 3.4. Estimation of interaction energy
coating surface. This shows that the fluorine atoms of the simulated
PVAc-PFO thin film/coating structures have the propensity to The interaction energy is a measure of interaction or adhesion
migrate to the surface, thereby, lowering the surface energy. The between the interface of simulated Al crystal and the PVAc-PFO
lower the surface energy the better the repellency of the surface systems. Based on the explanation provided in section 2.4 for
against the low surface tension oil-like molecules such as decane. estimating interaction energy, it can be stated that the more
This is in line with the experimentally observed phenomena by negative the value, the better is the adhesion the polymeric coating
Fabbri et al. [7] for perfluoropolyether-based organic-inorganic would have to the surface it is applied on. Fig. 8 illustrates the
hybrid coatings. variation of total interaction energy as a function of increasing PFO
The relative concentration of the fluorine atoms, R, has been content. All the energies obtained have been found to be negative in
found along the Miller plane having indices (h, k, l) as (1, 0, 0) with a sign indicating that all the formulated coatings show sufficient
bin width of 3. For determining the concentration profile, the atom binding to the surface.
coordinate information is taken as input and the relative concen- The interaction energy has a high negative (or less positive)
tration of the specified atoms in layers parallel to specific planes are value for lower PFO contents and approaches lowest negative (or
plotted. The relative concentration of fluorine atoms present along most positive) value for pure PFO based coating. This indicates that
the z-dimension is equal to the ratio of number density of such with increasing PFO content the interaction energy, and hence,
atoms in the considered volumetric bin or element to the average adhesion of the PVAc-PFO coating shall decrease as the PFO content
number density of such atoms in the entire system/amorphous cell. in the coating is increased. Hence, we see that the two parameters -
The maximum in the curve for simulated thin film explains that interaction energy of adhesion to applied surface and available free
all the fluorine atoms migrate to the two film side surfaces. The surface energy (estimated in section 3.2) work oppositely when
comparison with the curve for bulk coating indicates that the PFO content is varied in the PVAc based coating. This suggests that
propensity of fluorine atoms to migrate to the surface is higher in though increasing the PFO content would reduce the surface en-
the thin coating structure as compared to the bulk one, with the ergy (i.e. increases oleophobicity) as observed in section 3.2, the
average relative concentration of fluorine atoms being ~1.6 as PFO content in PVAc would need to be limited to maintain good
compared to 1.0 for the latter. In a real-time situation, one of the adhesion to the surface. Based on these two opposing factors, it is
thin film surfaces would get sandwiched with the surface to be proposed that the PFO content in PVAc be limited to 35.2e45.8% to
coated, and hence, accumulation of fluoro entities on that side have an optimized formulation with low surface energy for good
would be restricted. However, PFO due to its low molecular weight hydrophobicity/oleophobicity, and yet, high adhesion to the coated
(compared to other perfluoropolymers), would be inclined to easily surface.
diffuse to the other surface that is open to atmosphere.
Polymers containing atomic fluorine along the backbone or in
3.5. Contact angle measurements
the side chain possess many desirable properties, specifically lower
surface energies. This is due to the low polarizability and strong
We have used SPC/E (extended simple point charge) model for
electro negativity of the fluorine atoms leading to a strong dipole in
water with full Ewald summation for the calculation of its density
the CeF bond, as explained earlier by Pauling et al. [52] and Banks
and surface properties. The two layered systems obtained earlier
et al. [53]. Scheirs [54] explains that this could make fluoro-
from the last frame of MD run (section 2.3) are used to find out the
compounds simultaneously oleophobic and hydrophobic. In later
profiles of decane and water droplets measured in the direction
sections of this paper, we explore the level of oleophobicity and
perpendicular to the surfaces. Fig. 9(a), depicts the plot of the radius
hydrophobicity induced due to the fluorine in PFO, by estimating
of water droplet (on 35.2% PFO-PVAc system) and Fig. 9(b) repre-
the contact angles of decane and water over simulated surfaces.
sents the radius of decane droplet (on 35.2% PFO-PVAc system) as a
function of increasing distance from the coating surface of the
PVAc-PFO system (in the z-direction). The contact angle is then

Fig. 7. Relative Concentration of fluorine atoms (present in PFO molecules), R, verses

their spatial distance from the center of the amorphous cell in simulated bulk struc- Fig. 8. Illustration of variation of interaction energy, in mJ/m2, calculated from the MD
tures and thin film structures for 55.9% of PFO in PVAc system. and MM simulations as a function of weight % of PFO in PVAc.
202 N. Kumar, G. Manik / Polymer 100 (2016) 194e205

Fig. 9. Illustration of the procedure for determination of contact angles of: (a) water and (b) decane on 35.2% PFO-PVAc coating surface.

determined by plotting a tangent at x (Distance transverse in z- The surface profile of the decane molecules in Fig. 10(a)e(e)
direction) ¼ 0, i.e., the point of contact of the droplet and the illustrate that the contact angles of decane increase consistently
coating surface. The radii of these drops, r, are estimated (see with increased wt% of PFO. This is attributed to the increased
Appendix-1) through a direct proportionality relation (Eq. (6)) to oleophobicity of the simulated surface. For example, for decane on
the square root of relative concentration, R, (estimated in section coating, the contact angle increased from <10 to 73.32
, when the
3.3), with constant a estimated from simulation conditions. PFO content increases from 0 to 35.2 wt% in the simulated surface.
With an increase in PFO content, there is a decrease in solubility
r ¼ ðR*aÞ1=2 (6) parameter or surface energy that translates into higher water and

Fig. 10. Illustration of oil drop formation after 1000 ps simulations of decane molecules on the simulated coatings of (a) 0% PFO-PVAc (b) 12.4% PFO-PVAc (c) 24.1% PFO-PVAc (d)
35.2% PFO-PVAc (e) 74.7% PFO-PVAc and (f) 100.0% PFO systems.
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 203

Fig. 11. Illustration of water molecules after 1000 ps simulation on the simulated coatings of (a) 0% PFO-PVAc (b) 12.4% PFO-PVAc (c) 24.1% PFO-PVAc (d) 35.2% PFO-PVAc (e) 74.7%
PFO-PVAc and (f) 100.0% PFO systems.

oil repellency of the simulated surface or the thin film. both the hydrophobic and oleophobic properties of the PFO-PVAc
Likewise, the profile of the molecules in Fig. 11(a), (b), (c), (d) surface are significantly enhanced with increase in PFO content.
and (e) clearly depict that the contact angles of water also increase
as the wt% of PFO in the simulated coating increases. This is
attributed to increased hydrophobicity of the simulated surface. For 4. Summary and conclusions
example, for water on PVAc-PFO coating, the contact angle has
been found to increase from 58.27
to 84.76
when the PFO content We propose a PVAc-PFO based coating wherein PVAc provides
is varied from 0% (pure PVAc) to 35.2%. The simulated contact angle the film forming property and the PFO (a perfluoropolymer) pro-
of water on pure PVAc of 58.27
is very close to the reported value vides the essential oil repelling, and hence, dirt repelling property
of 60
(equilibrium contact angle) [55]. to the coating to maintain the surface stain-free. The present study
The results for contact angles for decane and water on the employs a molecular simulation strategy to estimate the properties
simulated coating structures for different PFO systems are tabu- and behavior of the proposed PVAc-PFO coating systems. Molecular
lated in Table 4. The contact angles for different simulation times mechanics and molecular dynamics approach has been applied to
have been estimated and it is observed that there is no significant generate equilibrated structures of such systems with varying PFO
change in the contact angles made by water droplet after 500 ps.
For example, on pure PVAc surface the contact angle made by a Table 4
simulated water droplet remained within 58
± 1
for simulations Contact angle simulation estimates for decane on coating and water on coating for
between 500 and 1000ps. Thus, 1000 ps. has been found to be different simulated PFO-PVAc systems.
sufficient to form stable liquid drops, and hence, provide stable and PFO (wt%) PFO:PVAc Water contact Decane contact
accurate contact angle estimates. It is interesting to note that the (chain ratio) angle (in degrees) angle (in degrees)
contact angle of water on PFO-PVAc system increases more or less
0.0 0\10 58.27
<10.00

linearly with increase in PFO content. However, the contact angle of 12.4 2\9 70.68
32.18

decane on the system increases swiftly from <10
to ~65
when 24.1 4\8 74.48
64.89

PFO content is changed merely by 24.1 wt% in contrast to the 35.2 6\7 84.76
73.32

104.21
81.20

relatively smaller change (73.32
e109.32
) when the PFO content is 55.9 10\5
74.7 14\3 121.18
94.51

varied from 35.2 wt% to 100 wt%. In general, this illustrates that 100.0 20\0 134.38
109.32

204 N. Kumar, G. Manik / Polymer 100 (2016) 194e205

content. A good agreement between simulation results for density, J. Phys. Chem. B 111 (10) (2007) 2431e2439.
[17] S.S. Jawalkar, T.M. Aminabhavi, Molecular modeling simulations and ther-
solubility parameter and surface energy calculations with the
modynamic approaches to investigate compatibility/incompatibility of poly(l-
literature data available for the pure bulk systems (PVAc and PFO) lactide) and poly(vinyl alcohol) blends, Polymer 47 (23) (2006) 8061e8071.
validates the accuracy of simulation protocol used. [18] S.S. Jawalkar, et al., Molecular modeling on the binary blend compatibility of
The surface energy of the simulated structures has been desir- poly(vinyl alcohol) and poly(methyl methacrylate): an atomistic simulation
and thermodynamic approach, J. Phys. Chem. B 109 (32) (2005)
ably found to decrease with increasing PFO content due to the 15611e15620.
surface aggregation observed with increasing PFO content. How- [19] B. Prathab, T.M. Aminabhavi, Atomistic simulations to compute surface
ever, the interaction energy of the formulated PVAc-PFO structures properties of poly(N-vinyl-2-pyrrolidone) (PVP) and blends of PVP/chitosan,
Langmuir 23 (10) (2007) 5439e5444.
with Al substrate (hypothesized as base coating substrate) [20] B. Prathab, T.M. Aminabhavi, Molecular modeling study on surface, thermal,
decreased with increasing PFO content. The opposite behavior mechanical and gas diffusion properties of chitosan, J. Polym. Sci. Part B
shown by surface energy and interaction energy limits the PFO Polym. Phys. 45 (11) (2007) 11.
[21] A. Dashevsky, et al., Physicochemical and release properties of pellets coated
content in PVAc to ~35.2e45.8 wt%. Based on above sets of with Kollicoat® SR 30 D, a new aqueous polyvinyl acetate dispersion for
reasoning, we propose a coating formulation of 35.2 wt% PFO in extended release, Int. J. Pharm. 290 (1e2) (2005) 15e23.
PVAc. Further, the contact angle estimates indicate that with [22] H. Proquitte
, et al., Breathing gas perfluorocarbon measurements using an
absorber filled with zeolites, Br. J. Anaesth. 91 (5) (2003) 736e738.
increasing PFO content the surface hydrophobicity showed an [23] Material Studio™, by Dassault Syste mes BIOVIA, UK (Accelrys®), http://
almost linear increase whereas oleophobicity showed a general accelrys.com/products/materials-studio/, (License purchased by IIT Roorkee).
increasing trend. If it is intended to develop a self-cleaning coating [24] D. Rigby, H. Sun, B.E. Eichinger, Computer simulations of poly(ethylene oxide):
force field, pvt diagram and cyclization behaviour, Polym. Int. 44 (3) (1997)
(a coating that cleans accumulated dirt if washed with water), then
311e330.
the acetate moieties of PVAc may be partially hydrolyzed to hy- [25] M, M.J.a.G., 1974. Journal Research of the National Bureau of Standards 78A.
droxyl groups. [26] T.J. Brice, R.I. Coon, The effects of structure on the viscosities of perfluoroalkyl
ethers and amines, J. Am. Chem. Soc. 75 (12) (1953) 2921e2925.
[27] S. Misra, P.D. Fleming, W.L. Mattice, Structure and energy of thin films of poly-
Acknowledgement (1,4-cis-butadiene): a new atomistic approach, J. Comp.-Aided Mater. Design
2 (2) (1995) 101e112.
The second author would like to sincerely acknowledge the [28] U. Natarajan, G. Tanaka, W.L. Mattice, Atomistic simulations of the surfaces of
thin films of random copolymers, J. Comp.-Aided Mater. Design 4 (3) (1998)
grant (Grant Code: FIG-400136-DPT/13-14) provided by Indian 193e205.
Institute of Technology Roorkee, through its MHRD FIG initiative, [29] D. He, D.H. Reneker, W.L. Mattice, Fully atomistic models of the surface of
for funding the work presented in this paper. The author gratefully amorphous polyethylene, Comput. Theor. Polym. Sci. 7 (1) (1997) 19e24.
[30] K.S. Katti, et al., Molecular interactions in intercalated organically modified
acknowledges the financial support provided. clay and clayepolycaprolactam nanocomposites: experiments and modeling,
Polymer 47 (1) (2006) 403e414.
Appendix A. Supplementary data [31] B. Prathab, V. Subramanian, T.M. Aminabhavi, Molecular dynamics simula-
tions to investigate polymerepolymer and polymeremetal oxide interactions,
Polymer 48 (1) (2007) 409e416.
Supplementary data related to this article can be found at http:// [32] I. Manassidis, A. De Vita, M.J. Gillan, Structure of the (0001) surface of a-Al2O3
dx.doi.org/10.1016/j.polymer.2016.08.019 from first principles calculations, Surf. Sci. 285 (3) (1993) L517eL521.
[33] W.R. Tyson, W.A. Miller, Surface free energies of solid metals: estimation from
liquid surface tension measurements, Surf. Sci. 62 (1) (1977) 267e276.
References. [34] R.S. Nelson, D.J. Mazey, R.S. Barnes, The thermal equilibrium shape and size of
holes in solids, Philos. Mag. 11 (109) (1965) 91e111.
[1] Huang, T.L.J., et al., Patterned article having alternating hydrophilic and hy- [35] H. Daoust, M. Rinfret, Solubility of polymethyl methacrylate and polyvinyl
drophobic surface regions. 1999, Google Patents. acetate, J. Colloid Sci. 7 (1) (1952) 11e19.
[2] R.H. Black, and D. Perlman, Improved anti-graffiti coatings and method of [36] P.A. Small, Some factors affecting the solubility of polymers, J. Appl. Chem. 3
graffiti removal. 1999, Google Patents. (2) (1953) 71e80.
[3] B. Prathab, et al., Molecular modeling and atomistic simulation strategies to [37] K.L. Hoy, J. Paint Technol. 42 (541) (1970) 2.
determine surface properties of perfluorinated homopolymers and their [38] P. EM, J. Polym. Sci. Polym. Lett. Ed. 15 (1) (1977) 1.
random copolymers, Polymer 47 (19) (2006) 6914e6924. [39] G. DiPaola-Baranyi, et al., Estimation of solubility parameters for poly(vinyl
[4] J.J. Reisinger, M.A. Hillmyer, Synthesis of fluorinated polymers by chemical acetate) by inverse gas chromatography, J. Chromatogr. A 166 (2) (1978)
modification, Prog. Polym. Sci. 27 (5) (2002) 971e1005. 349e356.
[5] B. Prathab, V. Subramanian, T.M. Aminabhavi, Computation of surface energy [40] M.K. Lindermann, Physical Constants of Polyvinyl Acetate, Polymer Handbook
and surface segregation phenomena of perfluorinated copolymers and blends 2nd Ed., J. Wiley, 1975, pp. V51eV55.
e a molecular modeling approach, Polymer 48 (1) (2007) 417e424. [41] T.K. Kwei, CRC Handbook of Polymer-liquid Interaction Parameters and Sol-
[6] R.V. Lakshmi, et al., Fabrication of superhydrophobic and oleophobic solegel ubility Parameters by Allan F M Barton, CRC Press, 1990. ISBN 0-8493-3544-2,
nanocomposite coating, Surf. Coat. Technol. 206 (19e20) (2012) 3888e3894. hard cover, 771 pp., price not available. Polymers for Advanced Technologies,
[7] P. Fabbri, et al., Perfluoropolyether-based organiceinorganic hybrid coatings, 1992. 3(1): pp. 47e47.
Polymer 47 (4) (2006) 1055e1062. [42] G.H. Gareth, P.E. Cassidy, J. Ken, D. Theodore, Fluoropolymers 2: Properties,
[8] C.E. Cansoy, U. Cengiz, The effect of perfluoroalkyl and hydrocarbon liquid Academic Publishers, New York, Boston, Dordrecht, London and Moscow,
chain lengths on oleophobic behaviors of copolymer surfaces, Colloids Surf. A 2002.
Physicochem. Eng. Aspects 441 (2014) 695e700. [43] Bart, J.C.J., Polymer Additive Analytics: Industrial Practice and Case Studies. 5.
[9] H. Sun, COMPASS: an ab initio force-field optimized for condensed-phase [44] R.-J. Roe, Papers presented at the 43rd National colloid SymposiumInterfacial
applications overview with details on alkane and benzene compounds, tension between polymer liquids, J. Colloid Interface Sci. 31 (2) (1969)
J. Phys. Chem. B 102 (38) (1998) 7338e7364. 228e235.
[10] A K., Bull. de l'Acad. Polonaise des Sci, 1962. 10 (11/12). [45] M.G. Freire, et al., Surface tension of liquid fluorocompounds, J. Chem. Eng.
[11] G. Polizos, et al., Scalable superhydrophobic coatings based on fluorinated Data 51 (5) (2006) 1820e1824.
diatomaceous earth: abrasion resistance versus particle geometry, Appl. Surf. [46] W.A. Zisman, Relation of the Equilibrium Contact Angle to Liquid and Solid
Sci. 292 (2014) 563e569. Constitution, in Contact Angle, Wettability, and Adhesion, American Chemical
[12] T.P.N. Nguyen, et al., Fabrication of superhydrophobic and highly oleophobic Society, 1964, pp. 1e51.
silicon-based surfaces via electroless etching method, Appl. Surf. Sci. 295 [47] M. Toselli, et al., Poly(ε-caprolactone)-poly(fluoroalkylene oxide)-poly(ε-cap-
(2014) 38e43. rolactone) block copolymers. 2, Therm. Surf. Prop. Polym. 42 (5) (2001)
[13] U. Cengiz, et al., Superhydrophobic terpolymer nanofibers containing per- 1771e1779.
fluoroethyl alkyl methacrylate by electrospinning, Appl. Surf. Sci. 258 (15) [48] M. Toselli, et al., Surface chemical analysis of poly(ε-caprolactone)e
(2012) 5815e5821. perfluoropolyetherepoly(ε-caprolactone) triblock copolymers by X-ray
[14] J. Bicerano, Prediction of Polymer Properties Third Edition, vol. Third, CRC photoelectron spectroscopy, Polym. Int. 52 (8) (2003) 1262e1274.
Press, New York, 2002. [49] RD, V.d.g., Eindhoven University of Technology, Eindhoven University of
[15] V.H. Dalvi, P.J. Rossky, Molecular origins of fluorocarbon hydrophobicity, Proc. Technology, The Netherlands, 2002.
Natl. Acad. Sci. 107 (31) (2010) 13603e13607. [50] J. Lüning, et al., Macromolecules (2001) 34.
[16] S.S. Jawalkar, et al., Molecular modeling simulations to predict compatibility [51] J. Lüning, D.Y. Yoon, J. Sto €hr, Importance of structural order for the low surface
of poly(vinyl alcohol) and chitosan Blends: a comparison with experiments, energy of perfluoroalkyl substituted polymethacrylates, J. Electron Spectrosc.
 N. Kumar, G. Manik / Polymer 100 (2016) 194e205 205

Relat. Phenom. 121 (1e3) (2001) 265e279. [54] J. Scheirs, Modern Fluoropolymers : High Performance Polymers for Diverse
[52] L. Pauling, The Nature of Chemical Bond, Cornell University Press, Ithaca, New Applications, Wiley, Chichester; Weinheim [u.a.], 2000.
York, 1960. [55] I.O. Ucar, et al., Combined XPS and contact angle studies of ethylene vinyl
[53] R. Banks, B.E. Smart, J. Tatlow, Organofluorine Chemistry Principles and Ap- acetate and polyvinyl acetate blends, Appl. Surf. Sci. 257 (22) (2011)
plications, Plenum Press, New York, 1994. 9587e9594.