Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Interfaces: Adsorption, Reactions, Films, Forces, Measurement Techniques, Charge Transfer, Electrochemistry, Electrocatalysis, Energy Production and Storage
Nanoscale Water Contact Angle on PTFE surfaces characterized by the MD-AFM imaging Jerzy Wloch, Artur Piotr Terzyk, Marek Wi#niewski, and Piotr Kowalczyk Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00257 • Publication Date (Web): 12 Mar 2018 Downloaded from http://pubs.acs.org on March 13, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Nanoscale Water Contact Angle on PTFE surfaces characterized by the MD-AFM imaging Jerzy Włoch1, Artur P. Terzyk*2, Marek Wiśniewski2, Piotr Kowalczyk3 [1] Faculty of Chemistry, Synthesis and Modification of Carbon Materials Research Group, Nicolaus Copernicus University in Toruń, Gagarin Street 7, 87-100 Toruń, Poland
[2] Faculty of Chemistry, Physicochemistry of Carbon Materials Research Group, Nicolaus Copernicus University in Toruń, Gagarin Street 7, 87-100 Toruń, Poland
[3] School of Engineering and Information Technology, Murdoch University, Murdoch 6150 WA, Australia
(*)Corresponding author: Artur P. Terzyk Tel: (+48) (56) 611-43-71, E-mail:
[email protected]
ABSTRACT: The aim of this study is to link polytetrafluoroethylene (PTFE) surface characteristics with its wetting properties in the nanoscale. To do this using Molecular Dynamics (MD) simulation three series of rough PTFE surfaces were generated by the annealing and compressing, and next characterized by the application of the MD version of
ACS Paragon Plus Environment
1
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 20
the Atomic Force Microscopy (AFM) method. The values of specific surface areas were additionally calculated. The TIP4P/2005 water model was used to the study of wetting properties of obtained PTFE samples. Simulated water contact angle (WCA) value for the most flat (but slightly rough) sample having PTFE density is equal to 106.94 ̊ and it is close to the value suggested for a perfect PTFE surface on the basis of experimental results. Also the changes in the WCA with PTFE compression are in the same range as experimentally reported. Obtained Molecular Dynamics simulation results make it possible to link, for the first time, the WCA values with the surface MD-AFM root mean square roughness and with PTFE density. Finally we show, that for PTFE wetting in the nanoscale the line tension is negligible and the Bormashenko's equation reduces to the Cassie - Baxter (CB) model. In fact our simulation results are close to the CB mechanism.
Keywords: contact angle, nanodroplets, PTFE, MD simulations INTRODUCTION
Wetting of fluorinated hydrocarbons surfaces has been often studied due to wide range of applications [1,2] However, considering the values of water contact angle (WCA) on polytetrafluoroethylene (PTFE [3]) reported in literature experimental data differ remarkably. Zisman et al. [4,5] reported the value of WCA equal to 108 ̊ and Tavana et al. [7] showed that the WCA on PTFE having the thickness in the range of 27-420 nm is equal to ca. 127 ̊ and almost does not depend on the thickness of the layer. Yekta-fard et al. [7] tabulated and compared the experimental WCA values (determined by different authors across the temperature range of T = 292-298 K in various atmospheres) measured on PTFE prepared under different conditions (pressed, polished, deposited etc.). The WCA values were across the range of 98-130 .̊ The authors also reported the results of in-house wetting experiment (T=298 K) on PTFE prepared without sanding, and on PTFE polished using SiC papers (the WCA value was equal to 130 ̊ and 121 ,̊ respectively). However, the most interesting are the results obtained for the samples pressed between the glass plates (using the pressures across the range of 103-883 bar). The WCA decreased with the pressing pressure (the lowest observed WCA value was equal to 112 ̊ ). The authors concluded, basing on the literature survey and on the results of in house experiments, that the WCA for a smooth, perfect PTFE surface should be close to 108 ̊ (and this is in agreement with our simulation results - see below).
ACS Paragon Plus Environment
2
Page 3 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Fluorinated hydrocarbons, especially PTFE, have been also widely applied to induce a surface hydrophobicity. For example Busscher et al. [8] prepared modified superhydrophobic fluoroethylenepropylene-PTFE surfaces. To increase hydrophobicity, the surfaces were ionetched and the WCA values were equal to 109 ± 2 ̊ for initial, and 140 ̊ for the etched surfaces. Van der Wal and Steiner [9] proposed an approach of superhydrophophobic surfaces preparation based on PTFE and polystyrene. The appearance of porosity on the surfaces increased the WCA value. For the reference PTFE sample it was determined as equal to 125 ± 2 ̊ and increased up to 171 ̊ due to the creation of pores. Recently Ruiz-Cabello et al. [10] prepared the PTFE-coated galvanized steel surfaces. Before coating, metal surfaces were roughened and the WCA values (across the range of 130-150 ̊) were dependent on the method of surface amorphisation. Few years ago Muthiah et al. [11] reported the application of electrospinning for the production of PTFE containing fibers having superhydrophobic properties. The observed WCA values were across the range of 144 ̊ - 165 ̊. From the above discussion one can conclude that not only the deposition of fluorinated hydrocarbons on a surface but also etching, creation of roughness and/or pores can rise a surface hydrophobicity. It is usually explained by the appearance of the Cassie-Baxter (CB) [12] effect. Considering PTFE, probably the most spectacular experimental evidence of the CB phenomenon was provided by Nilsson et al. [13] who developed a simple method of PTFE sanding, inducing superhydrophobic properties. Using sandpapers with different grit designations they created the CB states. The same group [14] proved that the sanding PTFE surfaces demonstrate drag reduction for the laminar flow of water. Also PTFE wrinkling can induce the CB effect, as it was reported by Scarratt et al. [15]. The authors showed that the WCA determined for the pure PTFE (127 ̊) can be remarkably increased, and for the wrinkled PTFE surfaces it is close to 169 .̊ Also Di Mundo et al. [16] studied different nano and micro textures of plasma modified surfaces of PTFE. The values of WCA were across the range of 160-175 ̊. However, there are also reports showing the decrease in WCA after PTFE irradiation. For example, Dasilva et al. [17] modified PTFE surface using vacuum UV photooxidation. The decrease in the WCA value from 115 ̊ to 70 ̊ was observed due to chemisorbed surface oxygen. Also Sabbatovskii et al. [18,15] reported the results of complex experimental studies of the WCA on plasma-modified PTFE AF samples. The WCA for the initial sample was equal to 120 ̊ and decreased exponentially with the time of plasma treatment. Simultaneously, WCA decreased linearly with the decrease in the F/C atomic ratio.
ACS Paragon Plus Environment
3
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 20
Molecular Dynamics (MD) simulation is very popular tool for the studying of wetting phenomena however, the results of water nanodroplets simulations on PTFE surface have been rarely reported. Up to our knowledge, the first MD simulation of water nanodroplet on the surface of PTFE was performed by Fan and Cagin [19]. The authors simulated 216 SPC/E water molecules, and the WCA was equal to 127 ̊ (it was larger than the experimental value quoted by the authors, equal to 108 ̊). They pointed out that the differences between simulation and experiment can be caused by an amorphous nature of real polymer surfaces and the presence of heteroatoms. The results of the first MD simulations of water droplets on amorphous polymer surfaces were published few years later by Hirvi and Pakkanen [20]. The authors proposed the method of amorphous surfaces preparation for the purpose of simulation studies. Following this method, Zhao and Cheng [21] simulated recently water droplets on PTFE surfaces. The WCA value was equal to 112.4 ̊. Few years earlier Dalvi and Rossky [22] applied the OPLSAA force field, for the simulation of WCA on different fluorinated surfaces. The authors performed MD simulation of wetting of the surfaces of hydro and fluorocarbon thiols self-assembled on Ag. The procedure of gradual transformation of hydrocarbons into fluorocarbons was used. The authors observed that the WCA values correlate with the Lennard-Jonnes (LJ) energy of the droplet-surface interactions. Additionally the authors proved that the differences in packing between hydro and fluorocarbons are responsible for the hydropobicity of fluorohydrocarbon surfaces. Recently Katasho et al. [23] presented very important results of WCA simulation on silica terminated with two types of surface groups. They concluded that water on CF3 containing surfaces is in the microscopic CB state. The exchange of CF3 groups into the CH3 leads to the appearance of micro Wenzel (W) states. From the experimental, as well as simulation reports described above it can be concluded that in the studies of WCA on PTFE some points need explanation/clarification. First of all we need to check the origin of the relation observed experimentally by Sabbatovskii et al. [18]. It is apparent that the rise in the F/C ratio should increase the WCA value [22,23]. However, in the real experiment etching is used for the changing of the surface F/C ratio. This leads to the creation of surface irregularities since a roughness appears. Both effects, i.e. the rise in the F/C ratio, and the appearance of PTFE surface roughness, cause the rise in the WCA value. However, during the real experiment both effects cannot be studied separately. Such possibility is offered by the MD simulation method. Thus to introduce a nanoscale roughness on PTFE surface at constant F/C ratio we use a simple method of a surface compressing and annealing.
ACS Paragon Plus Environment
4
Page 5 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Simultaneously it is possible to check how the compressing of PTFE surface changes the WCA value. To get the nanoscale insight into the mechanism of this polymer wetting it is necessary quantitatively describe its surface properties. Thus we propose in this study a new version of AFM images simulation [24] called the MD-AFM imaging method. This approach is based on the fundamentals of the Atomic Force Microscopy (AFM) and it is applied in the MD simulation of water on PTFE surfaces for the first time. Also the results of MD simulation of PTFE pressing and the study on the influence of surface heterogeneity on the WCA on PTFE are new and have not been reported yet. From the MD simulation results one can easily determine the state of water on a rough PTFE surface. Additionally the mechanism of PTFE wetting in the nanoscale can be studied, i.e. we check the applicability of the Wenzel (W) and CB models. This is important because today it is obvious, that the roughness not only in the micro, but also in the nanoscale plays a crucial role in wetting processes [25]. Here it is important to point out, that the results of our study provide input to the current debate about the application of macroscopic models to description of phenomena taking place at the nanoscale [26-28].We expect that our MD data will be important in the branches using PTFE to increase hydrophobicity of a surface.
MOLECULAR SIMULATIONS
PTFE structures. The OPLSAA force field for perfluoroalkanes tabulated by Watkins and Jorgensen [29] was applied for the preparation of the topology of C30F62 molecule. Next 33 structures (three series) were prepared to determine, for the first time, the influence of pressing and surface heterogeneity on the WCA values. The first series of samples was obtained using the compression performed for 0.3 ns in the N,V,T ensemble with the use of velocity Verlet integrator, at the temperature of T = 600 K controlled by the NoseHoover thermostat. To do this, 18 C30F62 molecules were placed between two repulsive walls, in the box having the dimensions x, y = 31.724 Å and z = 63.448 Å. The position of the top wall was gradually lowered in the z direction down to the value for which the density of the final (the most compressed) structure was equal to the density reported for pure PTFE (d = 2.15 g/cm3). The procedure of the creation of such samples was similar to this proposed by Hirvi and Pakkanen [20] and by Zhao and Cheng [21]. The densities of all 13 PTFE structures are tabulated in Table S1 (Supporting Information). The samples of this series are labeled as Tcom_i, where Tcom means compressed PTFE, and 21.2 Å + i is the distance (in Å) between the walls during the virtual compression process (Figure 1).
ACS Paragon Plus Environment
5
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 20
The second series (10 PTFE structures) was obtained by the heating of the PTFE layer. To do this, PTFE layer having the thickness equal to 21.2 Å was used. Next, the molecules located in the upper half-part of this layer (the thickness equal to 11.724 Å) were instantaneously heated to the temperature of T = 600 K and equilibrated for 150 ps. After this time all as prepared initial configurations were instantaneously frozen. The heterogeneity of the created structures was controlled by the distance between the surface and the repulsive wall (see Table S1 in Supporting Information). The samples of this series are labeled as Tann_i, where Tann_i means the annealed PTFE, and i is the distance between the walls (in Å) during the virtual annealing process (Figure 1). The third series (10 PTFE structures) differs if compared to the previous ones. No repulsive walls were used. The change of the annealing depth was used instead. The samples of this series are labeled as T2ann_i, where T2ann_i means annealed PTFE, and i is the annealing depth (in Å) (see Table S1 in Supporting Information and Figure 1).
Figure 1. Applied procedures of PTFE structures creation: compression (gray wall in the x,y plane; the creation of Tcom_i samples), annealing of the constant zone (the width marked by the blue surface in the y,z plane; the creation of Tann_i samples) and the gradually changed annealing depth in z direction (the annealing depths are schematically marked by yellow lines in the x,z plane; T2ann_i samples are created in this way).
Surface roughness estimation by MD-AFM method. To determine the roughness of the structures we propose a new version of AFM simulation method called the MD-AFM
ACS Paragon Plus Environment
6
Page 7 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
(Figure 2). In essence it is similar to molecular simulations - a tip scans the surface, generating its image (a map). Since all PTFE structures were kept rigid during simulations, AFM scanning was carried out in the same manner. All structure atoms were treated as an immobile LJ centers. The choice of the parameters of LJ equation is obvious - the same as during MD simulations. In the case of the scanning tip it is a bit more difficult. Apparently the water molecule should be used. However, the resulting image would be dependent on its orientation. To avoid this problem the LJ ball with the collision diameter (σ) equal to the kinetic size of the water molecule (2.65Å) is used instead. The scanning itself was carried out in the following way. In each step the tip was placed above the structure at the appropriate x,y position. Subsequently it is allowed to move downward along the z axis and it interacts with the structure. In this step MD formalism was used. When the tip meets the repulsive force the energy minimization between the tip and the nearest atom of the structure is calculated. This scanning step is ended when the position (and consequently distance to the nearest surface atom D in Figure 2) of the tip practically does not change, i.e. its position change falls below 10-5Å. After the energy minimization the position of the tip is stored and the tip is moved to the next x,y - position.
Figure 2. The basics of the MD - AFM: red and gray balls are PTFE atoms, blue ball is a tip (having the diameter of water molecule), D is a distance between a tip and the nearest surface atom, PD is the penetration depth of water nanodroplet sitting on the surface.
This procedure is repeated until the whole structure is scanned. In this way the AFMlike map is obtained. Subsequently from these maps five, essential parameters for the investigation of the wetting mechanism are calculated. The first one - Root Mean Square
ACS Paragon Plus Environment
7
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 20
Roughness (RMS) is borrowed from the real AFM experiment, as the parameter describing the standard deviation of the distribution of surface heights, given by [30]:
l
RMS =
1 n 1 2 { y( x)} dx = ∑ yi 2 ∫ l0 n i =1
(1)
where: l is the so called evaluation length on the x,y plane [30], n is the number of counts and y is the deviation of the position of the scanning tip from the mean value. RMS gives the information about the roughness of the wetted surface. The second, and obvious one is the top of the structure (blue dotted line in Figure 2). The third parameter is calculated from the combination of the MD-AFM map and a droplet isodensity profile obtained from the MD simulation of wetting. It is called by us the Penetration Depth (PD in Figure 2 and in Table S1 in the Supporting Information). It is the measure how deeply water molecules penetrate the PTFE structure. It is assumed that this is the distance between the bottom of the droplet and the top of the structure. It is applied for the calculation of the fourth important parameter - the fraction xsurf (calculated from MD-AFM maps, as the projections of the wetted surface part) of the surface being wetted. This parameter is necessary for the examination of the wetting mechanism if it is governed by the Cassie-Baxter equation (see below). Finally, the Structure Depth (SD) was also estimated. It is the deepest point penetrated by a tip in the MD-AFM experiment. The SD value is applied for the preparation of the MD-AFM images (Table S1 in Supporting Information).
Simulations of wetting. WCA and surface area values calculations. MD simulations of wetting were performed using the OpenMM 7.01 [31-35] - a high performance toolkit for MD simulation, supported by Phython 3.4.5. All calculations had been carried out on Titan GFORCE GTX Graphic Processing Unit. To study the WCA all structures were rigid during simulation. They were multiplied (4x4) and placed in the periodic simulation box having the dimensions x=y = 132 Å, and z=170 Å. As in our previous wetting studies [36-38] the TIP4P/2005 water model [39] was applied. The details of the WCA determination are provided in the Supporting Information. We also checked the influence of the initial configuration on the WCA value, studying 3 additional structures from the Tcom_i series. The differences between the WCA values determined for each structure were not larger than ± 0.5 ̊. For the calculation of the surface area values of all samples the double cubic lattice method proposed by Eisenhaber et al. [40] was used.
ACS Paragon Plus Environment
8
Page 9 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
RESULTS AND DISCUSSION
MD-AFM imaging of compressed PTFE surfaces and the WCA. Figures 3 and 4 collect the MD-AFM images of dry surfaces, and calculated from MD results images of wet surfaces together with water nanodroplet profiles on compressed PTFE samples (the series Tcom_i). The gradual compression of PTFE leading to the progressive rise in solid density (Table S1 in Supporting Information) causes simultaneous drop in the RMS values (eq 1) (Table S1 in Supporting Information). It can be seen that the most compressed sample (Tcom_0.0) having the density of pure PTFE (2.15 g/cm3) is almost flat (Figure 3A).
Figure 3. The MD-AFM images of the surfaces of compressed PTFE (Tcom_i) series (A) and the wetted part of surfaces calculated from the MD simulation of the wetting process (B).
ACS Paragon Plus Environment
9
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 20
Figure 4. The images of nanodroplets and the isodensity profiles in cylindrical coordinate system (A) and the x,y images for the droplets on compressed (Tcom_i) PTFE surfaces (B). Lines on the isodensity profiles: red - the mean value of the bottom part of the droplet used for the WCA calculations, the blue one - the position of the upper part of PTFE structure calculated from the MD-AFM scanning (see Figure 2).
At the same time the compression process causes the rise in the number of surface atoms having the contact with water nanodroplet molecules (Figure 3B). The WCA values decrease across the range of 117.15 ̊ - 106.94 ̊ for the Tcom_11.97 and Tcom_0.0 samples, respectively. It should be pointed out that the simulated WCA value for the most compressed PTFE sample is almost the same as estimated by Yekta-fard et al. [7] for a perfect PTFE surface (WCA = 108 ,̊ note that the surface Tcom_0.0 is not perfect, the RMS = 0.513 - see Figure 3A and Table S1 in Supporting Information). The nanodroplet profiles collected in Figure 4A and 4B clearly show that the droplet is almost perfectly spherical and the regular decrease in the WCA value during the virtual compression is caused by the rise in the density of surface atoms having a contact with nanodroplet. This is confirmed by the existence of a correlation between a cosine of WCA and PTFE density (Figure 5).
Figure 5. The dependence of the cosine of WCA on the density of compressed PTFE samples (Tcom_i).
Almost linear correlation shown in Figure 5 is new and has not been reported yet. However, it reflects the tendency of experimental data of Yekta-fard et al. [7] i.e. the linear relation between PTFE pressing pressure and the WCA. Moreover the experimental WCA values measured on pressed PTFE samples [7] are in similar range as observed during simulation, and shown in Figure 5. The appearance of this correlation is caused by the increasing number
ACS Paragon Plus Environment
10
Page 11 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
of PTFE atoms interacting with water molecules with the rise in the solid density. Similar effect was reported by Phillips and Dettre [41] who reported the decrease in WCA value with the rise in the thickness of fluoropolymers deposited on cover glass slides.
MD-AFM imaging of annealed PTFE surfaces and the WCA. Figures S1, S2, 6 and 7 collect the MD-AFM images of dry surfaces, and calculated from MD results images of wet surfaces together with water nanodroplet profiles on the surfaces of annealed PTFE. The changes in surface roughness (RMS - Tab.S1 in Supporting Information) are larger for the T2ann_i series due to applied procedure of in silico preparation (Figure 1). In this case the progressive rise of the annealing depth causes the appearance of strong surface roughness (Figure 6A and the values of the RMS - Table S1 in Supporting Information) resulting in progressive decreasing of water nanodroplet contact with the substrate (Figure 6B).
Figure 6. The MD-AFM images of the second series of the annealed PTFE (T2ann_i) surfaces (A) and the wetted part of them calculated from MD simulations of the wetting process (B).
ACS Paragon Plus Environment
11
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 20
Figure 7. The images of droplets and their isodensity profiles in cylindrical coordinate system (A) and the x,y images for the droplets on annealed (T2ann_i) PTFE surfaces (B). Lines on the isodensity profiles: red- the mean value of the bottom part of the droplet used for WCA calculations, blue onethe position of the upper part of PTFE structure calculated from the MD-AFM scanning.
Similar process of surface roughening is observed for the second annealed series (See Figure S1 and S2 in Supporting Information). The profiles of nanodroplets (Figures S2 and 7) are not so regular, especially for the droplet sitting on deeply annealed surfaces. One can also observe clearly visible decrease in water density around the surface fluoride atoms (Figures S2 and 7). As it was discussed by Dalvi et al. [22] and Katasho et al. [23] this effect of water density depletion causes the changes in water "cage structure" and the increase in a surface hydrophobicity.
Molecular insight into the Cassie - Baxter states of water nanodroplets on PTFE surfaces. The application of proposed in this study new version of the MD-AFM imaging makes it possible to check the influence of selected nanofactors characterizing PTFE surfaces on wetting mechanism. The results collected in Figure 8 show that the WCA values depend on the RMS (calculated by the use of MD-AFM procedure and eq 1, Table S1 in Supporting Information). With the rise in the RMS one can observe the rise in the WCA value, and this is often reported for the real PTFE surfaces after the creation of microscopic-sized roughness see for example [13].
ACS Paragon Plus Environment
12
Page 13 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Figure 8. The correlation between the RMS calculated from the MD-AFM data using eq 1 and the cosine of water contact angle on studied PTFE samples.
To check the mechanism of wetting on PTFE surfaces at the nanoscale the equation proposed by Bormashenko [42] was tested. This equation takes into account the line tension [43-49] effect, and was developed for the wetting of rough heterogeneous surfaces. What is interesting, Bormashenko's approach leads to the equation that, at the special conditions, can be reduced to equations proposed by Marmur, Miwa et al. [50,51] and Wong and Ho [52]. The final version of Bormashenko's equation is:
cos(WCA) B = RFxsurf cos(WCA)Y + xsurf − 1 −
τ γ LV
1 ξ + a
(2)
where (WCA) B is the contact angle on a rough surface for a nanodroplet having the line tension τ, γ LV is the liquid-vapor surface tension, ξ is the perimeter of the triple line per unit area of the substrate under the droplet.
To use eq 2 it is necessary to check the influence of line tension on the WCA values. To do this, the same procedure as previously [37] was applied, namely we studied the WCA values for droplets containing different number of water molecules: 1000, 2000, 4000, 6000 and 8000. Two PTFE samples were chosen: Tcom_11.97and Tcom_0.0 because they have extremely different densities (see Tab. S1 in Supporting Information). The results collected in Figure S3 (Supporting Information) show practically no influence of droplet size on the WCA, i.e. the WCA change is within the estimated calculation error of ca. 0.5 ̊ . These results are consistent with those presented by us in [37] where it was shown, that for low potential energy of droplet - substrate interactions the line tension effect is negligibly small. Besides in
ACS Paragon Plus Environment
13
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 20
each series the reciprocal value of the perimeter radius (necessary for line tension calculations) of the contact line varies within the very narrow range (0.27 - 0.35 nm-1). Thus the line tension effect on studied surfaces has no influence on the WCA values.
Before we discuss the applicability of Bormashenko’s model (eq 2) in the nanoscale, in Figure 9 we present the results of interesting correlation between the RMS and RF -1. One can observe that the both values are linearly correlated with the line passing through the origin. This correlation is somewhat unexpected since one should find such a correlation when comparing the RMS values with the total surface accessible to water. It occurred however, that such clear correlation in this case is not observed.
Figure 9. The correlation between the RMS calculated from MD-AFM using eq 1 and the roughness factor of studied PTFE samples.
Figure 10. Graphical representation of the fitting of Cassie - Baxter (A) and Wenzel (B) models to simulation data at the nanoscale (dashed line shows the perfect fit x=y), blue dots - the cos(WCA) for hypothetical fully flat surface.
The origin of this correlation needs further studies and we leave the explanation of this phenomenon to the readers. But in our opinion the above observation can make in future the
ACS Paragon Plus Environment
14
Page 15 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
link between the experimental AFM measurements and the mechanism of wetting on rough surfaces. In Figure 10 we show the comparison of WCA values calculated from eq 2 . Note that if the line tension is small, at the specific conditions eq 2 reduces to W [53] and/or CB [12, 50] models that can be applied using the surface parameters calculated by our MD-AFM method, and the method proposed in [40] (see Supporting Information, and Fig. S4 for details). It should be stressed that the only fitted parameter, and the same for all series is the hypothetical cos(WCA)Y equal to -0.25 (104.48°). It can be seen that the process of PTFE wetting at the nanoscale is closer to the CB, and not the W mechanism. However, it should be pointed out that the partial correlation between the W model and simulation results is observed for the highest cosine values (Fig. 10B). It is not unique. For this region almost complete wetting occurs (see Figs. 3B, 6B and S1B). In this case CB equation reduces to the W one. Our MD simulation results are in line with experimental data. Nilsson et al. [13] as well as van der Wal and Steiner [9] suggested the existence of water in CB states on rough and PTFE surface, while Muthiah et al. [11] on PTFE coated nanofibres. Scarratt et al. [15] concluded that on wrinkled PTFE surfaces water is in "partially collapsed and/ or suspended CB state". This conclusion can explain why the linear plots in Figure 11A are not perfect. However, to check the applicability of other modifications of CB models for MD data fitting additional simulations should be performed. This will be the subject of our future studies.
CONCLUSIONS
The results of systematic studies, using Molecular Dynamics (MD), on the influence of PTFE surface roughness on the wetting by water are reported for the first time. The new version of the MD-AFM method offers the possibility of the calculation of a surface structural parameters that can be easily applied in future studies for linking surface heterogeneity with its adsorption and wetting (as well as with the other) properties. Our simulation data lead to almost the same WCA value as suggested in experiment for a perfect PTFE surface (106.94 ̊). We also observe the decrease in the WCA with pressing, i.e. the same behavior as reported in real experiment. The decrease in the WCA with PTFE pressing is strictly related to the rise in PTFE density. For the most rough surfaces water nanodroplet density profiles show drop around surface fluoride atoms, and this is the reason of PTFE hydrophobic properties. Calculated with use the MD-AFM method RMS values determine the WCA values in the nanoscale. Finally we show, that for PTFE wetting in the nanoscale the line tension is ACS Paragon Plus Environment
15
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 20
negligible and the Bormashenko's equation reduces to the CB model. This model was applied for MD data using the parameters from our MD-AFM method. Up to our knowledge this is the first simulation proof of the Bormashenko’s model applicability in the nanoscale.
ASSOCIATED CONTENTS Supporting Information The Supporting Information is available free of charge on the ACS Publications website.
ACKNOWLEDGEMENT The authors gratefully acknowledge financial support by the Polish National Science Centre (NCN) grant OPUS 13 UMO-2017/25/B/ST5/00975.
REFERENCES (1) Tavana, H.; Simon, F.; Grundke, K.; Kwok, D.Y.; Hair, M.L.; Neumann, A.W. Interpretation of contact angle measurements on two different fluoropolymers for the determination of solid surface tension. J. Coll. Interf. Sci. 2005, 291, 497-506. (2) Diaz, M.E.; Savage, M.D.; Cerro, R.L. The effect of temperature on contact angles and wetting transitions for n-alkanes on PTFE. J. Coll. Interf. Sci. 2017, 503, 159-167. (3) Jarvis, N.L.; Zisman, W.A. Surface chemistry of fluorochemicals, 1965, Oct. NRL Report, p.3 (4) Fox, H.W.; Zisman, W.A. The spreading of liquids on low energy surfaces. I. Polytetrafluoroethylene. J. Coll. Interf. Sci. 1976, 56, 251-254. (5) Ellison, A. H.; Fox, W.; Zisman W.A. Wetting of fluorinated solids by hydrogen-bonding liquids. J. Phys. Chem. 1953, 57, 622-627. (6) Tavana, H.; Petong, N.; Hennig, A.; Grundke, K. Contact angles and coating film thickness. J. Adhes. 2005, 81, 29-39. (7) Yekta-fard M.; Ponter, A.B. Surface treatment and its influence on contact angles of water drops residing on Teflon and copper. J. Adhes. 1985, 18, 197-206. (8) Busscher, H.J.; Stokroos, I.; Van der Mei, H.C.; Rouxhet, P.G.; Schakenraad, J.M.; Preparation and characterization of superhydrophobic FEP-Tefon Surfaces. J. Adhes. Sci. Technol. 1992, 6, 347-356. (9) Van der Wal, P.; Steiner, U. Super-hydrophobic surfaces made from Teflon. Soft Matt.
2007, 3, 426-429.
ACS Paragon Plus Environment
16
Page 17 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
(10) Ruiz-Cabello, F.J.M.; Amirfazli, A.; Cabrerizo-Viılcheza, M.; Rodrıguez-Valverdea, M. A. Fabrication of water-repellent surfaces on galvanized steel. RSC Adv. 2016, 6, 7197071976. (11) Muthiah, P.; Hsu, S.H.; Sigmund, W. Coaxially electrospun PVDF-Teflon AF and Tefoln AF-PVDF core-sheath nanofiber mats with superhydrophobic properties. Langmuir
2010, 26, 12483-12487. (12) Cassie, A.B.D.; Baxter, S. Large contact angles of plant and animal surfaces. Nature
1945, 155, 21-22. (13) Nilsson, M.A.; Daniello, R.J.; Rothstein, J.P. A novel and inexpensive technique for creating superhydrophobic surfaces using Teflon and sandpaper. J. Phys. D: Appl. Phys.
2010, 43, 045301. (14) Song, D.; Daniello, R.J.; Rothstein, J.P. Drag reduction using superhydrophobic sanded Teflon surfaces. Exp. Fluids, 2014, 55, 1783 - 1788. (15) Scarratt, L.R.J.; Hoatson, B.S.; Wood, E.S.; Hawkett, B.S.; Neto, C. Durable superhydrophobic surfaces via spontaneous wrinkling of Teflon AF. ACS Appl. Mater. Interf.
2016, 8, 6743-6750. (16) Di Mundo, R.; Bottiglione, F.; Palumbo, F.; Notarnicola, M.; Carbone, G. Filamentary superhydrophobic Teflon surfaces: Moderate apparent contact angle but superior air-retaining properties. J. Coll. Interf. Sci. 2016, 482, 175-182. (17) Dasilva, W.; Entenberg, A.; Kahn, B.; Debies, T.; Takacs, G. A. Surface modification of Teflon PFA with vacuum UV photo-oxidation. J. Adhes. Sci. Technol. 2006, 20, 437-455. (18) Sabbatovskii, K. G.; Dutschk, V.; Nitschke, M.; Simon, F.; Grundke, K. Properties of the Teflon AF1601S surface treated with the low-pressure argon plasma. Koll. Zhurn. 2004, 66, 239-247. (19) Fan, C. F.; Cagin, T. Wetting of crystalline polymer surfaces: A molecular dynamics simulation. J. Chem. Phys. 1995, 103, 9053-9061. (20) Hirvi, J.T.; Pakkanen, T.A. Molecular dynamics simulations of water droplets on polymer surfaces. J. Chem. Phys. 2006, 125, 144712. (21) Zhao, L.; Cheng, J. Analyzing the molecular kinetics of water spreading on hydrophobic surfaces via molecular dynamics simulation. Nat. Sci. Rep. 2017, 7, 10880. (22) Dalvi, V.H.; Rossky, P.J. Molecular origins of fluorocarbon hydrophobicity. PNAS,
2010, 107,13603-13607.
ACS Paragon Plus Environment
17
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 18 of 20
(23) Katasho, Y.; Liang, Y.; Murata, S.; Fukanaka, Y.; Matsuoka, T.; Takahashi, S. Mechanisms for enhanced hydrophobicity by atomic-scale roughness. Nat. Sci. Rep. 2015, 5, 13790. (24) Konuyama, M.; Kazuya, T.; Tazawa, K.; Hirotani,, A.; Yamano, H.; Kubo, M.; Broclawik, E.; Miyamoto, A. Simulation of AFM/LFM by molecular dynamics: role of lateral force in contact-mode AFM imaging. Surf. Sci. 1996, 357-358, 222-227. (25) Jadhav, A.D.; Ogale, S.B.; Prasad, B.L.V. Carbon nano horn and bovine serum albumin hierarchical composite: towards bio-friendly superhydrophobic protein film surfaces. J. Mater. Chem. 2008, 18, 3422-3425. (26) Miyahara, M.; Tomohisa, Y.; Nakamura, J.; Okazaki, M. Simple Evaluation Scheme of Adsorbate-Solid Interaction for Nano-Pore Characterization Studied with Monte Carlo Simulation, J. Chem. Engn. Jpn. 2000, 33, 103-112. (27) Peterson, B.K.; Gubbins, K.E. Phase transitions in a cylindrical pore, Mol. Phys. 1987, 62, 215-226. (28) Takei, T.; Chikazawa, M.; Kanazawa, T. Validity of the Kelvin equation in estimation of small pore size by nitrogen adsorption, Coll. Polym. Sci., 1997, 275, 1156-1161. (29) Watkins, E.K; Jorgensen, W.L. Perfluoroalkanes: Conformational analysis and liquidstate properties from ab initio and Monte Carlo calculations. J. Phys. Chem. A 2001, 105, 4118-4125. (30) Gadelmawla, E.S.; Koura, M.M.; Maksound, T.M.A.; Elewa, I.M.; Soliman, H.H. Roughness parameters. J. Mater. Process. Technol. 2002, 123, 133-145. (31) Friedrichs, M.S.; Eastman, P.; Vaidyanathan, V.; Houston, M.; LeGrand, S.; Beberg, A.L.; Ensign, D. L.; Bruns, C.M.; Pande, V.S. Accelerating molecular dynamic simulation on graphics processing units. J. Comp. Chem. 2009, 30, 864-872. (32) Eastman, P.; Pande, V.S. Constant constraint matrix approximation: a robust, parallelizable constraint method for molecular simulations. J. Chem. Theor. Comput. 2010, 6, 434-437. (33) Eastman, P.; Pande, V. S. Efficient nonbonded interactions for molecular dynamics on a graphics processing unit. J. Comput. Chem. 2010, 31, 1268 - 1272. (34) Eastman, P.; Pande, V.S. OpenMM: A hardware-independent framework for molecular simulations. Comput. Sci. Engn. 2010, 12, 34-39. (35) Eastman, P.; Friedrichs, M.S.; Chodera, J.D.; Radmer, R.J.; Bruns, C.M.; Ku, J.P.; Beauchamp, K.A.; Lane, T.J.; Wang, L.-P.; Shukla, D.; Tye, T.; Houston, M.; Stich, T.; Klein, C.; Shirts, M.R.; Pande, V.S. OpenMM 4: A reusable, extensible, hardware
ACS Paragon Plus Environment
18
Page 19 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
independent library for high performance molecular simulation. J. Chem. Theor. Comput.
2013, 9, 461-469. (36) Włoch, J.; Terzyk, A.P.; Kowalczyk, P.; Korczeniewski, E.D.; Kaneko, K. Switchable hydrophobicity/hydrophilicity of a HOPG surface - Comment on the paper by Y. Wei and C.Q. Jia, Carbon, 87 (2015) 10-17. Carbon 2017, 115, 571-573. (37) Włoch, J.; Terzyk, A.P.; Kowalczyk, P.; New forcefield for water nanodroplet on a graphene surface. Chem. Phys. Lett. 2017, 674, 98-102. (38) Włoch, J.; Terzyk, A.P.; Wiśniewski M.; Kowalczyk, P. Nanoscale insight into the mechanism of a highly oriented pyrolytic graphite edge surface wetting by “interferencing” water. Langmuir 2017, 33, 8562-8573. (39) Abascal, J.L.F.; Vega, C. A general purpose model for the condensed phases of water: TIP4P/2005. J. Chem. Phys. 2005, 123, 234505. (40) Eisenhaber, F.; Lijnzaad, P.; Argos, P.; Sander, C.; Scharf, M. The double cubic lattice method: Efficient approaches to numerical integration of surface area and volume, and to dot surface contouring of molecular assemblies. J. Comput. Chem. 1995, 16, 273-284. (41) Phillips, R.W.; Dettre, R.H. Application of ESCA and contact angle measurements to studies of surface activity in a fluoropolymer mixture. J. Coll. Interf. Sci. 1976, 56, 251-254. (42) Bormashenko, E. General equation describing wetting of rough surfaces. J. Coll. Interf. Sci. 2011, 360, 317-319. (43) Pethica, B.A. The Contact Angle Equilibrium. J. Coll. Interf. Sci. 1977, 62, 567-569. (44) Boruvka, L.; Neumann, A.W. Generalization of the Classical Theory of Capillarity. J. Chem. Phys. 1977, 66, 5464-5476. (45) Drelich, J. The Significance and Magnitude of the Line Tension in Three-Phase (SolidLiquid-Fluid) Systems. Col. Surf. A. 1996, 116, 43-54. (46) Pompe, T.; Herminghaus, S. Three-Phase Contact Line Energetics from Nanoscale Liquid Surface Topographies. Phys. Rev. Lett. 2000, 85, 1930-1933. (47) Widom, B. Line Tension and the Shape of a Sessile Drop. J. Phys. Chem. 1995, 99, 2803-2806. (48) Pompe, T. Line Tension Behavior of a First-Order Wetting System. Phys. Rev. Lett.
2002, 98, 076102-1. (49) Amirfazli, A.; Neumann, A.W. Status of the Three-Phase Line Tension: A Review. Adv. Coll. Interf. Sci. 2004, 110, 121-141. (50) Marmur, A. Wetting on hydrophobic rough surfaces: To be heterogeneous or not to be? Langmuir 2003, 19, 8343-8348.
ACS Paragon Plus Environment
19
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 20
(51) Miwa, M.; Nakajima, A.; Fujishima, A.; Hashimoto, K.; Watanabe, T. Effects of the surface roughness on sliding angles of water droplets on superhydrophobic surfaces. Langmuir 2000, 16, 5754-5760. (52) Wong, T.S.; Ho, Ch.M. Dependence of macroscopic wetting on nanoscopic surface textures. Langmuir 2009, 25, 12851-12854 (53) Wenzel, R.N. Resistance of solid surfaces to wetting by water. Ind. Engn. Chem. 1936, 28, 988-994.
ACS Paragon Plus Environment
20
