Experimental and Dissipative Particle Dynamics ... - ACS Publications

Jan 6, 2015 - Escuela Superior de Ingeniería Química e Industrias Extractivas, Instituto Politécnico Nacional, Unidad Profesional “Adolfo López. Mateo...
0 downloads 0 Views 5MB Size
Article pubs.acs.org/Langmuir

Surface Photografting of Acrylic Acid on Poly(dimethylsiloxane). Experimental and Dissipative Particle Dynamics Studies David Ramírez-Gutiérrez,†,§ Carlos Nieto-Draghi,† Nicolas Pannacci,† Laura V. Castro,‡ Fernando Á lvarez-Ramírez,§ and Benoit Creton*,† †

IFP Energies nouvelles, 1 et 4 avenue de Bois-Préau, 92852 Rueil-Malmaison, France Escuela Superior de Ingeniería Química e Industrias Extractivas, Instituto Politécnico Nacional, Unidad Profesional “Adolfo López Mateos”, Zacatenco, 07738, México City, México § Instituto Mexicano del Petróleo, Eje Central Lázaro Cárdenas 152, 07730, México City, México ‡

ABSTRACT: This work includes both experimental and theoretical studies of the wetting property changes of water on a surface of poly(dimethylsiloxane) (PDMS) modified with different amounts of acrylic acid (AA). The default surface properties of PDMS were changed from hydrophobic to hydrophilic behavior which was characterized with contact angle measurements by two approaches: (i) experimental tests of samples subjected to a photografting polymerization procedure to obtain a functionalized surface and (ii) DPD (dissipative particle dynamics) simulations which also involve the calculation of sets of repulsive parameters determined following two methods: the use of the “Blends” module in the Materials Studio software and the calculation of cohesive energy density with molecular simulations. Changes of contact angle values observed from both experimental and numerical simulation results provide qualitative and quantitative information on the wetting behavior of photografted surfaces.



INTRODUCTION There has been extensive research on the control of wetting behavior of materials exhibiting strong hydrophilic and hydrophobic surfaces.1 Such property tuning of chosen materials proved to be interesting for numerous potential applications, including biomaterials,2 microelectronics, surface coatings,3 and microfluidics devices fabrication.4 Microfluidics experimental devices have been developed as miniaturized systems aiming to decrease experimental time and to reduce sample/reagent consumption5 for tasks such as drug screening, catalyst screening, clinical diagnostic, genetics research, chemical synthesis, cellular analysis, and biodetection. Enhancement of phenomena due to high surface/volume ratio and small Reynolds number of flows on a 100 μm scale has led to great interest in microsystems for numerous physical chemistry studies of complex fluids. As an example, petroleum-related experiments based on microfluidic devices have been used to study carbon dioxide diffusivity in bitumen,6 separation of hydrocarbon and polar fractions from crude oil,7 and asphaltene content and carboxylic acid content determination in crude oil,8,9 or for modeling porous rock structures in enhanced oil recovery studies with microfluidic channels.10 The wetting behavior of microfluidic channels is controlled by both chemical structure and surface roughness of selected polymer-based substrates that is critical for successful applications. Polymeric materials are preferred for the fabrication of microfluidic devices, and the materials used © XXXX American Chemical Society

include polymers such as poly(methyl methacrylate) (PMMA), polystyrene (PS), polyimide (PI), branched polyethylene (PE), polycarbonate (PC), poly(ethylene terephthalate) (PET), and poly(dimethylsiloxane) (PDMS).11 All these polymer substrates must have chemical inertness or resistance regarding the different environments to which they will be exposed. PDMS is a very attractive material for the prototyping of microfluidic channels due to its low cost related to its commercially availability in addition to its easy fabrication and manipulation techniques. PDMS also offers optical transparency, durability, high flexibility, and good thermal and chemical stability. PDMS has favorable advantages required for biological and medical applications, such as nontoxicity, biocompatibility, low biological activity, and mechanical properties similar to those of human soft tissue.11,12 However, for most applications which are endowed with these beneficial properties, the natural and high hydrophobicity of PDMS is a disadvantage, due to its low surface energy. PDMS structural composition, which involves methyl groups on the surface, render the material hydrophobic and chemically inert.13 Several techniques have been developed for manipulating surface properties and preserving bulk properties, to change the default hydrophobicity that is inappropriate for most Received: September 16, 2014 Revised: January 5, 2015

A

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Scheme 1. Mechanism of Grafting Polymerization Reaction

hydrophilic behavior changes. As well as modifying wetting behavior, new functional groups introduced on a bulk PDMS substrate are used as precursors for reactions that lead to versatile platforms that can be modified by reacting systems utilizing functional groups already attached on the surface.26 Maleic anhydride (MA) copolymers allow reactions for introducing further functionalities and/or immobilization of bioactive molecules,16 but it is difficult to attach MA directly on PDMS by chemical treatment as mentioned previously; MA can be covalently bonded by reacting with NH2 groups on the PDMS surface as a result of low-pressure ammonia plasma treatment. Carboxylic acid-functionalized films, from AA grafting, are interesting platforms for chemical modification to obtain amide and ester linkages.13 Besides experiments, it is important to complement observations with numerical simulations to investigate interactions occurring during surface modifications, i.e., from hydrophilic to hydrophobic character or reverse. Atomistic level simulations27,28 such as Monte Carlo29 (MC) and Molecular Dynamics30 (MD) provide reliable results as long as accurate intermolecular potentials are used. Despite the constant growth of computational resources, molecular scale methods remain restricted for practical applications due to accessible time and length scales. Very recently, Yiapanis et al. have investigated wetting properties of functionalized poly(ethylene glycol) at nanoscale level using a computational approach, and contact angle values were determined considering the contact area between the surface and the droplet and the height of the droplet.31 Xu et al. studied wetting behavior on functionalized surfaces using MD simulations.32 In the case of hydrophobic surfaces (CH3, CHCH2, OCH3), water droplets conserve spherical shape while for hydrophilic surfaces (CN, NH2, and COOH), water molecules spread all over the surface, emphasizing the role of hydrogen-bonding effects. Theoretically and experimentally obtained results have proven that carboxylic functional groups present on a functionalized surface promote hydrophilic behavior and wetting of water droplets as a consequence. This is due to strong and more probable hydrogen bonding between COOH groups and water.33,34 An alternative methodology is dissipative particle dynamics (DPD), which is a mesoscopic scale simulation technique that uses coarse-grained particles (beads) to represent clusters of atoms or molecules.35 DPD can be seen as an extension of MD simulations to higher time and length scales, and an interesting candidate to study complex fluids36−38 and wetting effects.39−41 DPD simulations are usually used to qualitatively describe systems, but quantitative comparisons between experiments and DPD simulation results remain scarce. Hereafter, we describe an innovative use of DPD

applications. Surface changes to enhance properties can be classified in two types of modifying methods: physical and chemical modifications. Physical modification refers to the fabrication of structures that change roughness, using soft lithography technology, an electron beam in physical vapor deposition, and alter shapes on the substrate surface14 at microand nanoscales. Chemical modifications consist of the introduction of active functional groups to the molecular structure of PDMS by reactions that take place at the surface.14 Plasma grafting and surface modification with polymers are the two main methods utilized for chemical modification. Plasma treatment is usually reliable, reproducible, and relatively inexpensive, which introduces functional groups onto the surface of PDMS. The etching effect and surface oxidation by plasma treatment, using different gases such as Ar, O2, CO2, ethanol,15 NH3, and air,16 allow the generation of silica-like layers. The plasma method also permits immobilization (crosslinking) of hydrogel layers and polymeric films predeposited on the substrate at low pressure.17 Functionalized surfaces subjected to the plasma procedure have short effective lifetimes, due to reorientation of nonpolar groups in the bulk toward the surface, or surface hydrophilic groups toward the bulk.16,18 Surface modification with polymers is another method for chemical modification commonly used in PDMS treatment to graft monomers to the surface and comprises activation of the surface and monomer chain grafting. This strategy employs high energy treatments; plasma treatment was proposed in this way,19,20 but ultraviolet (UV) light is mostly used for PDMS functionalization. UV radiation provides the required energy to activate the surface and can promote polymerization (with the appropriate photoinitiator) and propagation of vinyl- and allyltype monomers (Scheme 1). The combination of photoinitiator and UV radiation generates free radicals on the methyl groups at the PMDS surface which are required for initiation and subsequent propagation reactions in graft polymerization; benzophenone (Ph2CO) is widely used as the photoinitiator.21−23 Chain transfer agents are typically used in this process; benzyl alcohol (BnOH) is used to limit growing chains of polymers.24 The result of the process in Scheme 1 is the formation of attached polymer brushes with end-functionalized groups on the surface, changing the hydrophobic behavior depending on the type of monomer used in the process. Monomers used in photografting treatments contain functional groups such as OH, COOH, and NH2 for different applications; monomers such as acrylic acid (AA), acrylamide (AM), dimethylacrylamide (DMA), 2-hydroxyethyl acrylate (HEA), 2-hydroxyethyl methacrylate (HEMA), and poly(ethylene glycol)-monomethoxyl acrylate (PEG) have been studied24,25 as homopolymers and copolymers grafted on the PDMS surface to determine B

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

on the air-side surface of the photografted PDMS samples at room temperature, (ii) and using a blunt syringe needle, a droplet of deionized water was placed on photografted PDMS samples immersed in a silicone oil media. The silicone oil was purchased from the VWR International Company, and the reported silicone oil viscosity is 500 cSt at 25 °C. Measurements of contact angles were realized using a binocular microscope (Olympus) connected to a camera PL-B741U (Pixelink). The image treatments were performed using ImageJ software version 1.43u. For each exposure time and intensity couple, five photografted PMDS samples were prepared and contact angles measured. Simulation Details. DPD, which was first introduced by Hoogerbrugge and Koelman,47 is a mesoscale computational modeling method used to simulate hydrodynamic behavior of complex fluids systems, consistent with Navier−Stokes equations. DPD is based on the molecular dynamics principle in which interacting particles (beads) with uniform mass (m), position (ri), and velocity (vi) evolve according to classical Newtonian equations of motion:

simulation results leading to quantitative predictions of contact angle values. In this work, we propose a novel investigation of surface modifications for PDMS pieces involved in microfluidic channels by measurements of contact angles using both experiments and DPD simulations. The following section presents details about (i) performed surface modifications (photografting polymerization with AA), (ii) experiments, and (iii) DPD simulations. The third section proposes a discussion about two approaches to parametrize DPD simulations: obtained experimental data compared with DPD simulation results. This paper ends with a fourth section, which contains our conclusions.



MATERIALS AND METHODS

Experimental Section. Substances under consideration in this study were purchased and employed without additional purification. The poly(dimethylsiloxane) support was prepared using the Sylgard 184 (a two-part poly(dimethylsiloxane)), a silicone elastomer kit purchased from Dow Corning. The Sylgard 184 resin was mixed with its curing agent (cross-linking catalyst) in 9:1 (w/w) proportion, during 5 min. The 9:1 ratio is not the one recommended by Dow Corning (10:1) but was chosen to follow an internal protocol for microfluidics applications. Indeed, this ratio enables the increase in the degree of cross-linking in the PDMS network and results in an enhancement of the elastomer stiffness.42,43 The obtained mixture was transferred to a Petri dish and polymerized at 70 °C during 1.5 h. After the curing process, the elastomeric PDMS was cut into approximately rectangular parallelepiped pieces with the following dimension: 20 mm × 20 mm × 5 mm. The photografting polymerization procedure was carried out using a solution containing a photoinitiator (solution A), and a solution containing a monomer (solution B). Solution A consists of 10 wt % benzophenone dissolved in a mixture 35:65 (wt %/wt %) of acetone and distilled water, respectively, resulting in a transparent solution after sufficient agitation. Benzophenone was purchased from SigmaAldrich in certified purity greater than 99%. The acetone was purchased from VWR International Company in certified purity greater than 99.5%. The anhydrous (99%) acrylic acid (AA) purchased from Sigma-Aldrich Company was dissolved in water to obtain solution B having a 10 wt % concentration of AA.44 Practically, PDMS pieces are immersed in solution A during 1 min. Pieces are then removed from solution A and rinsed under distilled water for discharging the benzophenone that was not absorbed onto the PDMS surface. This procedure ensures reactive sites on the surface. It is necessary to rinse PDMS pieces after the immersion process, since nonadsorbed benzophenone parasitic polymerization could happen on the bulk of AA solution.45 PDMS pieces are then fixed in a crystallizing dish (which must be clean to ensure sufficient adhesion between surfaces), and the dish is filled with solution B to immerse PDMS pieces. This whole ensemble is exposed to ultraviolet (UV) radiation. The spotlight source used was a Hamamatsu LC8 Lightningcure L9566 with a high intensity mercury−xenon lamp, and the distance between the lamp and the sample is kept fixed at 20 cm. A discussion about variations in the UV exposure time and intensity, which define the hydrophobic or hydrophilic character of the treated surface, is proposed in the following section. Finally, the surface is washed from residual monomers using distilled water, and then photografted PDMS samples are dried. The contact angle has been used to describe the wetting capability of a liquid to a solid,46 being a measurement of hydrophobic/ hydrophilic character of a surface,24 and is defined as the angle formed from tangential lines of fluid−fluid and fluid−solid interfaces.40 Static water contact angle measurements at air-side with values ≥90° describe a hydrophobic behavior or a partial wetting. Contact angles between a water droplet and photografted PDMS surfaces having various hydrophobic or hydrophilic characters were experimentally determined for two systems: (i) placing a droplet of deionized water

dri = vi dt fi = m

dvi = dt

(1)

∑ (FCij + FijD + FRij + FSij) (2)

i≠j

where fi is the interaction force exerted on bead i by other particles. The force fi can be described as the summation of forces that includes stochastic and nonstochastic pairwise bead interaction terms. Forces are generally denoted as conservative, FCij , dissipative, FDij , random, FRij , and spring, FSij. This summation of forces is spread over all neighboring particles within a cutoff distance (rc). The conservative force is a linear weight function as follows:

⎧ ⎪ a ij(1 − rij/rc)riĵ (rij < rc), FCij = ⎨ ⎪ 0 (rij ≥ rc) ⎩

(3)

where rij is the relative distance between beads i and j, and r̂ij is the unit vector defined as r̂ij = (ri − rj)/|rij|. The term aij is the maximum repulsion between particles i and j. Conservatives forces could involve many types of interactions, such as electrostatic. Repulsion parameters aij represent the maximum potential required and is described as a soft repulsive interaction. Two ways to determine these parameter values are discussed hereafter. FDij and FRij forces act between pairs of beads, dissipative force could be interpreted as the modeling of viscous forces, γ is the friction factor, and the random force is related to the term σ, that defines the random thermal fluctuations in the fluid.48 Thus, the combined effects result from a special thermostat which conserves momentum, leading to a correct description of hydrodynamics.49,50 These forces are defined in eqs 4 and 5.

FijD

FRij

⎧− γωD(r )(r ̂ v )r ̂ (r < r ), ⎪ ij ij ij ij ij c =⎨ ⎪ 0 (rij ≥ rc) ⎩

(4)

⎧ σωR (r )θ r ̂ (r < r ), ⎪ ij ij ij ij c =⎨ ⎪ 0 (rij ≥ rc) ⎩

(5)

where vij = vi − vj, the difference of velocities between beads i and j. ωD and ωR are weight functions with dependence to the relative distance between particles and are related by the following equations:

σ 2 = 2γkBT

(6)

rij ⎧ (r < rc), ⎪1 − rc ij ω (rij) = [ω (rij)] = ⎨ ⎪ (rij ≥ rc) ⎩ 0 D

C

R

2

(7) DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir θij =

ξij

silicone oil used in experimental measurements are both composed of the same substance; the difference resides in the coarse-grained structures. While solid PDMS is a cross-linked polymer, which was represented in our DPD simulations as fixed and randomly distributed (B) beads, the silicone oil is composed of short chains of PDMS; this coarse-grained structure is formed with (B) beads in a linear chain arrangement. Though solid and liquid phases are chemically PDMS, beads are renamed as indicated in Table 1. The molecular structure of repeating units of PDMS is used to calculate the repulsion parameters of (B) particle. A bead is defined essentially in terms of the chemical constitution of the system. The bead is a representation that conserves aspects from the initial structure. The pair repulsion parameters, aij, were calculated between each pair of molecular structures. The calculation of these parameters is related to Groot’s approximation53 that utilizes Flory− Huggins parameter χij to determine the maximum repulsion between the particle pair i and j, eq 12.

(8)

Δt

The term θij in eq 8 is a randomly fluctuating variable and takes the following form with ξij as a random number drawn from a uniform distribution with zero mean and unit variance ⟨ξij⟩ = 0, ⟨ξ2ij⟩ = 1. Δt is the time step used for the resolution of the Newton equations of motion.51 Finally, for mesomolecules containing more than one bead in its structure, there are bonded interactions described by Hooke’s law. Spring force c, in the expression of eq 9, is associated with the strength of the bond interaction, and rs is the distance of the bond. Recently, Deguillard et al. observed a non-negligible effect of c and rs parameter values on interfacial tension, for a water/surfactant/oil system at large surfactant concentration.52 In this work, DPD simulations were performed using rs = 5 Å and c = 5 kcal/mol/Å2.

FSij = − c(rs − rij)riĵ

(9)

a ij = a ii + 3.27χij for ρ = 3

Coarse-Graining Level and DPD Parameters Description. There are three different components involved in the system: water, PDMS, and AA. Calculations of repulsion parameters were made using molecular structures (Table 1); for AA and PDMS, repeating units of

(12)

The term aii is the self-repulsion interaction that is derived from the compressibility of pure water (aii = 75kBT/ρ = 25, with kBT = 1 and ρ = 3). In this work Flory−Huggins parameters χij were determined by two procedures: Blends module54,55 and Molecular Dynamics (MD) simulation to obtain the cohesive energy density (CED) at 298 K. Blends module54,55 evaluates the energy of interaction (Eij) between two species, and in conjunction with the coordination number (Zij) it is possible to calculate a mixing energy (Emix) of particles i and j using eq 13.

Table 1. Coarse-Grained Structures for the Components Involved in Our DPD Systems

Emix =

1 [Z ij(E ij)T + Zji(Eji)T − Z ii(E ii)T − Zjj(Ejj)T ] 2

(13)

This evaluation of energies is made by an excluded-volume constraint method. Two characteristic molecular structures are represented by their van der Waals surfaces: one takes the role of base molecule that keeps static while a screen molecule will surround the base molecule until both van der Waals surfaces no longer overlap each other. At this configuration, the energy is calculated and stored in a histogram. This process is successively applied to different configurations. The coordination number is related to the number of screen molecules that surround a base molecule until it is completely saturated. The Flory−Huggins parameters are then derived from Emix. The second method involves the calculation of the Flory−Huggins parameter by an approximation using solubility parameters (δi) of the pure components with the following formula:

χij =

the two species were chosen. For both solid PDMS and silicone oil, the same molecular structure (DPD bead) was used, which means the same repulsion parameter in liquid and solid PDMS-built regions. An additional bead type (D) is considered in this study to mimic a nonpenetrable solid. This type of bead is fictional and is not associated with any atomistic structure. The role of this bead, or rather of the conservative force associated with this bead, is to establish a barrier to prevent passage of other beads. Keeping beads (D) fixed with high aij values gives, as a result, a region that emulates a solid. To construct the model mesomolecules in the simulations, scales must be established. The coarse-graining level is defined by grouping three molecules of water in one bead (DPD particle); this way, length and time scales, respectively, are related to the following equations:49

R c = 3.107(ρNm)1/3

(10)

τ = (14.1 ± 0.1)Nm5/3

(11)

V (δi − δj)2 kBT

(14)

where V is the volume of a bead (with all beads of the same average volume R3c /ρ).56 The solubility parameters of species i and j are obtained by the cohesive energy density (CED), δ = (CED)1/2. The CED is simply the cohesive energy per unit of volume. The solubility parameters of pure systems are obtained from atomistic simulations such as MD. In particular, for our case study, MD methodology includes the construction of three periodic boundary cells, each one containing 20 molecules of corresponding substances: water, PDMS, and AA. The simulation cells of pure systems have volumes of 598.29, 7855.45, and 2460.22 cubic angstroms, respectively. In both the Blends and MD-CED techniques, the calculation of the Flory−Huggins parameter χij was carried out using the COMPASS force field.57−59 COMPASS is a general all-atom force field for atomistic simulation of common organic molecules and polymers, developed by using state of the art ab initio and empirical parametrization techniques, and it has been validated for a wide variety of systems. In our calculations we used the charge assignment by the force field for the interatomic interactions. The CED was obtained by performing NpT ensemble molecular dynamics simulations of the previously constructed water, PDMS, and AA cells with P = 0 and T = 298 K. Berendsen’s barostat with decay

where Nm is the number of water molecules in one DPD particle. The spatial and temporal ranges in all DPD simulations is Rc = 6.4633 Å, and the time step equals 88 ps, when Nm = 3 and ρ = 3. Generally, coarse-grained structures are formed, delimiting the amount of atoms contained in a particle. Water and AA particles are each represented by beads (A and C), respectively. Solid PDMS and D

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 1. Description of the simulation box composition. In this illustration the size of B beads for silicone oil has been reduced to visualize the water droplet. constant 0.1 ps and Nose’s thermostat with Q ratio 1 were used to control pressure and temperature, respectively. The system is MDrelaxed during 1 × 105 simulations steps using a time step of 1 fs. Configurations were then collected every 2500 steps until completing 40 100-ps configurations. The CED is calculated as the average of the CED from 40 output configurations. Flory−Huggins parameters obtained by Blends and MD-CED methodology were utilized for calculating the interaction parameter value for each pair of particles. With the matrix of pairwise repulsion interaction and dissipation strength γ = 4.5, in reduced units, the parameter scheme is basically completed. DPD Simulations. This work is focused on the study of the contact angle changes of a water droplet on a surface of PDMS, varying the graft density of AA on it, in particular the case of an immersed system in silicone oil, as described in the Experimental Section. Therefore, a cubic simulation box with 100 Å side lengths is constructed. Each simulation box contains 10 990 beads so that the density of all systems is set to 3 in reduced units. In DPD simulations, to represent the effects of a solid wall, fixed particles are usually used to represent a solid matrix.40 All particles located at solid layer regions were constrained on its motion, by fixing positions. The simulation box mimicking the fully grafted surface system is composed of fixed center of mass (CoM) of DPD beads in different slabs (slabs 2, 3, and 4, see Figure 1) of average density of 3 in reduced units. In slab 4, we have placed a support layer (100 × 100 × 23.5) composed of nonpenetrable beads at the bottom of the simulation box. Above this latter slab, we have placed a layer (slab 3) of PDMS beads (100 × 100 × 4.5). Then slab 2 (100 × 100 × 2), which is composed of CoM of individual DPD beads that can be either AA (bead C) or PDMS (bead B), is placed at the top of the fixed region. The composition of slab 2 was adjusted from fully hydrophobic to fully hydrophilic by changing the ratio of PDMS and AA beads to mimic different grafting levels. A water droplet is placed on top of slab 2, and the remaining empty space (slab 1) was filled by silicone oil mesomolecules. Each region was independently relaxed and thermalized before the system was assembled to ensure appropriate initial conditions for simulations. The description of the grafted surface is quite simple compared to experiments; however, it is enough to catch the main characteristics of hydrophobic/hydrophilic surfaces. The DPD calculations carried out in this work were performed using the Mesocite60 module as implemented in Materials Studio 7.0.55

on the PDMS surface. However, intensity and time variables are related to energy. This means that, just as time exposure is increased, the energy delivered to the surface also increases; therefore, there is a direct dependency of energy but not only on time exposure; this may be due to a required energy to activate the surface and begin the polymerization reactions. To determine this dependency of energy, a series of PDMS pieces were treated by fixing a value of energy and changing UV irradiation conditions of intensity and exposure time. We chose to keep the energy (E) fixed at 2911.77 mJ/cm2, which corresponds to an intensity (ϕ) of 9.71 mW/cm2 with an exposure time (t) of 300 s. Such parameters were taken as starting conditions to be affected by a linear factor n (eq 15) to change the intensity and time exposure but keeping the same amount of energy applied to the surface. E = ϕt =

⎛ϕ⎞ ⎜ ⎟(nt ) ⎝n⎠

(15)

Five conditions of intensity and exposure time couples were defined by five different values for the linear factor, i.e., n = 0.5, 0.75, 1.0, 1.5, and 2.0. For instance, when n = 2, the exposure time is 600 s and the intensity is 4.85 mW/cm2. Five samples of PDMS pieces were treated according to each condition for UV irradiation, and contact angle measurements were performed. Figure 2 shows that for a fixed value of energy but varying intensity and exposure time conditions according to the “n” factor, measured contact angle values remain roughly constant. Contact angle values lie in between 39.88° and 41.50°, which



RESULTS AND DISCUSSION The grafting of various monomers on to Sylgard 184 PDMS surfaces has already been studied by Hu et al.24 The authors showed for all considered monomers an increase of the graft density with UV irradiation time, with a low initial rate of grafting attributed to the required time to form a sufficient number of radicals on the PDMS surface. Thus, during our experiments the intensity of the lamp was fixed and the exposure time was increased to modify the grafting level of AA

Figure 2. Measured contact angles of a water droplet in air on treated PDMS surfaces for five values of linear factor, i.e., n = 0.5, 0.75, 1.0, 1.5, and 2.0. Each error bar results from average of 5 independent measurements. E

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

similar to the minimum contact angle reached by Hu et al. for PDMS surfaces grafted with AA, indicating that the AA is deposited at the highest density on the PDMS surface.24 Hu et al. reached this fully grafted surface with a UV exposure time of 3.5 h (using benzyl alcohol as reagent) while we obtained a treated surface having similar properties after a treatment with UV irradiation during 300 s using benzophenone. Activation of the surface or initiation and propagation polymerization reactions occur more rapidly under our conditions. Water Droplet in Dense Media. The samples tested under atmospheric conditions (in air) presented previously have shown that the amount of AA grafted on the surface can be tuned, controlling the UV intensity and/or time exposure variables during the photografting process. The samples were tested in a silicone oil media by immersing the PDMS pieces and placing a water droplet on the surface, and the measurement was made once the water had spread (Figure 5a). The measurements obtained with this test show a decrease in the contact angle value just as the time of exposure increases, similar to the measurements under air medium. However, a comparison between both series of measurements indicates that in the oil media, the contact angle tendency is displaced to higher values (Figure 6). Contact angle values lie in between 77.54° and 140.48°. Another series of measurements was performed with oil media, utilizing the Sylgard 184 resin without a curing agent in order to use the silicone resin in the liquid state. Figure 5b shows images of this latter system which was inverted up and down in the measurements, due to the higher density of the silicone resin as compared to water. Comparing the results between the two types of oil media, the tendency of reducing the contact angle is approximately the same, but the values obtained for the silicone resin are higher than that of silicone oil. These changes can be explained by differences in the water/ silicon surface tension for the two types of liquid silicones. Indeed, silicone oil is described as a short linear chain of PDMS and the silicone resin is a longer chain of PDMS. To discuss these wetting behaviors, we measured interfacial tensions for our systems. We performed pendent drop measurements using a Teclis Tracker tensiometer. We obtained for the silicone resin media (i.e., part A of the PDMS without part B to keep the liquid state) the following values: γsilicone resin−air = 18.0 mN/m and γsilicone resin−water = 40.0 mN/ m. For the silicone oil media, we obtained the following values: γsilicone_oil−air = 20.9 mN/m and γsilicone_oil−water = 39.5 mN/m. Uncertainties of ±0.3 mN/m and ±3.0 mN/m are expected for surface tension measurements of the silicone−air and silicone− water, respectively. Tap water surface tension was found to be 72 mN/m ± 0.1 mN/m. Equations 16 and 17 can be obtained considering Young’s equation applied to a sessile water drop (w) on a solid PDMS surface (s) in two media: (i) the air (a) with the contact angle θ1 (Figure 3), (ii) the silicone oil (o) with the contact angle θ2 (Figure 5a) with obvious notations on the surface tensions γij between i and j phases.

indicates that the contact angle value is directly related to the amount of energy applied to the surface. Water Droplet in Air Medium. PDMS pieces were irradiated utilizing 30% of the maximal intensity of the UV lamp that corresponds to a surfacic flux 9.71 mW/cm2 and with a fixed distance between lamp and samples of 20 cm. Samples were exposed to UV radiation during different times to vary the amount of grafted AA on the surface of PDMS; such times were 50, 200, 300, and 500 s. For each exposure time, five PDMS samples were treated and tested. Modified PDMS samples were tested by placing a deionized water droplet on the treated surface and measuring contact angles, noting that five untreated PDMS pieces were also tested. Figure 3 shows the images of

Figure 3. Pictures of a water droplet in air on treated PDMS surfaces (t = 0, 50, 200, and 300 s). The picture corresponding to t = 500 s is not shown, as it is similar to that obtained when t = 300 s.

tested surfaces, which were obtained with a binocular microscope connected to a camera. Contact angle values were determined by tracing tangential lines on water−PDMS and water−air interfaces, and so-obtained values are presented in Figure 4. The contact angle of the nontreated PDMS is

Figure 4. Measured contact angles for a water droplet in air on treated PDMS surfaces for UV exposure times of 0, 50, 200, 300, and 500 s. The dashed line stands for the plateau value obtained by Hu et al. (45°).24

110.38°, which corresponds to a hydrophobic behavior. This latter value is in excellent agreement with that reported by Hu et al.24 As the UV exposure time increases, water droplets spread over the surface due to higher amounts of grafted AA, which confers a hydrophilic characteristic to the surface. After 300 s of UV irradiation, the measured contact angle value is 44.23° and a similar value is observed for t = 500 s. This value is

cos θ1 =

cos θ2 = F

γsa − γsw γwa

(16)

γso − γsw γwo

(17) DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 5. Pictures of water droplets immersed in a (a) silicone oil media and in a (b) silicone resin media, on treated PDMS surfaces (t = 0, 50, 200, and 300 s).

bead type is used to mimic both the surface and the dense media. Once components of systems studied by DPD simulations are identified, it is required to obtain the repulsion parameters in pairs. As mentioned previously, the repulsion parameters were calculated using the Flory−Huggins parameters, which were determined from two methods: the Blends module in the Materials Studio software55 and calculations of the cohesive energy density (CED) using Molecular Dynamics. The sets of parameters obtained from these two latter methods are presented in Table 2. The binary repulsion parameter is the maximum repulsion between two particles i and j. It is a dimensionless quantity that can be interpreted as an indirect measurement of the affinity between pairs of particles; the closer to the self-repulsion value (aii = 25), the less repulsion particles have with each other. Values of repulsion parameters presented in Table 2 show that the AA beads (C) have a hydrophilic character as compared to the PDMS bead which is described as hydrophobic. These trends are observed for the two sets of parameters. While the character of the repulsion is similar for the PDMS−water interaction, the magnitude of the parameter obtained from CED is approximately twice that from the Blends module. However, the solubility parameters found with CED give values similar to the experimental ones. A comparison between the calculated and experimental solubility is given in Table 3. Contact Angle Values from DPD Simulations. Once all interactions are defined in the system, the simulation boxes can be constructed. As previously described and illustrated in Figure 1, the simulation boxes are composed of various slabs. Slab 3 contains PDMS particles (bead B), and slab 2 is filled with AA particles (bead C) until the volume is completely saturated (ρ = 3). These two layers are in a constrained condition, by fixing the position of the involved particles during simulations. A sphere of water particles (bead A) is placed in the remaining volume, and the outer part of the sphere is filled

Figure 6. Evolution of the contact angle value of a water droplet in oil media on PDMS surfaces using various exposure times (t = 0, 50, 200, and 300 s).

With the surface tension being mainly linked to the surface chemistry, we can expect surface tensions involving the nontreated solid PDMS phase (s) similar to the ones obtained considering liquid PDMS (i.e., silicone resin). Thus, we assume that for the 0 s exposure time, γsa ∼ γsilicone resin−air, and γsw ∼ γsilicone resin−water, and for the same physical chemistry argument, we can consider γso ∼ 0 mN/m. Combining experimental interfacial tension values with this elementary hypothesis leads to a contact angle θ1 ∼ 107°, and contact angles θ2 ∼ 180°, for both silicone resin and silicone oil. The calculated contact angles are in good agreement with the θ1 = 110.38° experimental value in air and the nonwetting tendency of the water droplet shown when immersed in oil media (silicone resin: θ2 ∼ 160°, and silicone oil: θ2 ∼ 140°). Repulsion Parameter Calculations. For the construction and performance of DPD simulations, only the lighter media, i.e., silicone oil, was considered. The system studied through DPD simulations contains water, AA, and PDMS, and as described in the section dedicated to simulation details, PDMS

Table 2. Binary Repulsion Parameter (aij) Obtained from the Blends Module in the Materials Studio Software and Calculations of the Cohesive Energy Density (CED)a Blends module A B C D a

cohesive energy density

A

B

C

D

A

B

C

D

25.00 57.80 36.77 300

57.80 25.00 61.23 300

36.77 61.23 25.00 300

300 300 300 25.00

25.00 108.52 39.90 300

108.52 25.00 52.86 300

39.90 52.86 25.00 300

300 300 300 25.00

A to D denote DPD bead types as presented in Table 1. G

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Table 3. Comparison between Experimental and Calculated Solubilitiesa compound

δ = (CED)1/2

experimental

water PDMS AA

22.7 6.5 15.9

23.461 7.361 12.362

a

The simulated solubility values were obtained by averaging the CED of each of the 40 output configurations from the NpT-MD. Both simulated and experimental solubilities are in expressed in (cal/ cm3)1/2.

with silicone oil coarse-grained structures. The so-obtained simulation box represents the system describing a fully grafted surface of PDMS with AA.The number of AA particles that saturate the surface was used as a reference to determine different percentages of AA grafted, to vary the amount of AA on the surface. On the basis of energy profiles, an initial period of 100 000 steps was left for equilibration, and the simulation productions were carried out for an additional 900 000 steps. Drop configurations were extracted on the last 500 000 DPD steps at every 25 000 steps. For each frame, the measurement of the contact angle was performed in the same way as during experiments: each frame was used as an image and subjected to ImageJ software to obtain the contact angle. With this procedure, an evolution of the contact angle with increasing amounts of AA on the surface can be observed (Figure 7). We

Figure 8. Comparison of simulated contact angle values of a water droplet in a silicone oil medium on PDMS supports with different amounts of AA, with corresponding experimental values. A double abscissa scale is used: dark symbols follow the dark x-axis (bottom) and gray symbols follow the gray x-axis (top).

values. This comparison is based on the assumption that, at 300 s of exposure time, the resultant treated surface is fully grafted with AA. This assumption is based on results shown in Figure 3 which indicates, approximately, the same value of contact angle with exposure time greater than 300 s. Thus, a fully grafted surface in DPD simulations can be correlated to a treated surface of PMDS with an energy of 2911.77 mJ/cm2, and a direct comparison between experimental and DPD simulation results such as in Figure 8 can be proposed. The plotted tendency drawn from DPD simulation results is close to experimental contact angle values.



CONCLUSIONS Using a two-step photografting polymerization procedure (photoinitiator absortion and UV irradiation) to modify a PDMS surface with a hydrophilic monomer (AA), we are able to change wetting properties of the surface, tuning from hydrophobic to hydrophilic behavior. To characterize the behavior of the treated surface, we measured contact angle values for a water droplet on treated surfaces. We showed that the amount of AA grafted on the surface has a dependency on the UV energy applied to the system during the photografting process. Due to the energy that is defined as the product of intensity and time of exposure, we are able to use these variables to control the level of grafting. Measurements of treated surfaces made in an air medium demonstrated a decrease of the contact angle as the time of UV exposure increases, obtaining a minimal value of approximately 45° consistent with values found in the literature. We also measured contact angle values for a water droplet on the same treated surfaces placed in oil media. We also proposed a direct comparison between these latter experimental results with numerical results obtained from dissipative particle dynamics simulations. The DPD method allows us to develop a series of simulations to quantify the wetting property of water on a surface of PDMS with different amounts of grafted AA contained in an oil media. Two approaches were investigated to determine bead−bead repulsive parameters used in DPD simulations: the use of the Blends module in the Materials Studio software and calculations of the cohesive energy density (CED), resulting in two sets of parameters. The amount of AA particles on a layer of PDMS particles was varied with different

Figure 7. Snapshots of DPD simulations of water droplets (blue) on PDMS (violet) for different amounts of AA (pink). These simulations were performed with repulsion parameters calculated with Blends methodology. In all the AA concentrations, the silicone oil beads have been removed for clarity.

have also tested size effects, increasing the size of our systems by 1.5. The comparison of the two system sizes has shown no dependence of the contact angle values with respect to the droplet size. Figure 8 contains the corresponding measured values for each 10% of grafted AA on the surface, for the two sets of parameters obtained in this work. Surprisingly, Figure 8 indicates that contact angle values computed for a water droplet on an untreated PDMS surface are similar for the two sets of parameters while the repulsive interaction between water and PDMS beads obtained from CED is twice that from Blends. On the contrary, the small differences observed in repulsive parameter values between AA and silicon oil (PDMS) beads and AA and water beads lead to distinct contact angle values as the surface becomes fully grafted. Figure 8 also proposes a comparison between experimental and simulated contact angle H

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(16) Cordeiro, A. L.; Zschoche, S.; Janke, A.; Nitschke, M.; Werner, C. Functionalization of poly(dimethylsiloxane) surfaces with maleic anhydride copolymer films. Langmuir 2009, 25, 1509−1517. (17) Nitschke, M.; Zschoche, S.; Bailer, A.; Simon, F.; Werner, C. Low pressure plasma immobilization of thin hydrogel films on polymer surfaces. Surf. Coat. Technol. 2004, 185, 120−125. (18) Tóth, A.; Bertóti, I.; Blazsó, M.; Bánhegyi, G.; Bognar, A.; Szaplonczay, P. Oxidative damage and recovery of silicone rubber surfaces. I. X ray photoelectron spectroscopic study. J. Appl. Polym. Sci. 2003, 52, 1293−1307. (19) Barbier, V.; Tatoulian, M.; Li, H.; Arefi-Khonsari, F.; Ajdari, A.; Tabeling, P. Stable modification of PDMS surface properties by plasma polymerization: Application to the formation of double emulsion in microfluidic systems. Langmuir 2006, 22, 5230−5232. (20) Massey, S.; Duboin, A.; Mantovani, D.; Tabeling, P.; Tatoulian, M. Stable modification of PDMS surface properties by plasma polymerization: Innovative process of allylamine PECVD deposition and microfluidic devices sealing. Surf. Coat. Technol. 2012, 206, 4303− 4309. (21) Schneider, M. H.; Willaime, H.; Tran, Y.; Rezgui, F.; Tabeling, P. Wettability patterning by UV-Initiated graft polymerization of poly(acrylic acid) in closed microfluidic systems of complex geometry. Anal. Chem. 2010, 82, 8848−8855. (22) Abate, A. R.; Krummerl, A. T.; Lee, D.; Marquez, M.; Holtze, C.; Weitz, D. A. Photoreactive coating for high-contrast spatial patterning of microfluidic device wettability. Lab Chip 2008, 8, 2157− 2160. (23) Li, W.; Nie, Z. H.; Zhang, H.; Paquet, C.; Seo, M.; Garstecki, P.; Kumacheva, E. Screening of the effect of surface energy of microchannels on microfluidic emulsification. Langmuir 2007, 23, 8010−8014. (24) Hu, S.; Ren, X.; Bachman, M.; Sims, C. E.; Li, G. P.; Allbritton, N. Surface modification of poly(dimethylsiloxane) microfluidic devices by ultraviolet polymer grafting. Anal. Chem. 2002, 74, 4117−4123. (25) Hu, S.; Ren, X.; Bachman, M.; Sims, C. E.; Li, G. P.; Allbritton, N. L. Tailoring the surface properties of poly(dimethylsiloxane) microfluidic devices. Langmuir 2004, 20, 5569−5574. (26) Zhang, J.; Chen, Y.; Brook, M. A. Facile functionalization of PDMS elastomer surfaces using thiol−ene click chemistry. Langmuir 2013, 29, 12432−12442. (27) Allen, M. P.; Tildesley, D. J. Computer simulation of liquids; Oxford Science Publications: Oxford, U.K., 1987. (28) Gubbins, K. E.; Moore, J. D. Molecular modeling of matter: Impact and prospects in engineering. Ind. Eng. Chem. Res. 2010, 49, 3026−3046. (29) Theodorou, D. N. Progress and outlook in Monte Carlo simulations. Ind. Eng. Chem. Res. 2010, 49, 3047−3058. (30) Maginn, E. J.; Elliott, J. R. Historical perspective and current outlook for Molecular Dynamics as a chemical engineering tool. Ind. Eng. Chem. Res. 2010, 49, 3059−3078. (31) Yiapanis, G.; Maclaughlin, S.; Evans, E. J.; Yarovsky, I. Nanoscale wetting and fouling resistance of functionalized surfaces: A computational approach. Langmuir 2014, 30, 10617−10625. (32) Xu, Z.; Song, K.; Yuan, S. L.; Liu, C.B. Microscopic wetting of self-assembled monolayers with different surfaces: A combined molecular dynamics and quantum mechanics study. Langmuir 2011, 27, 8611−8620. (33) Moussa, S. G.; McIntire, T. M.; Szori, M.; Roeselová, M.; Tobias, D. J.; Grimm, R. L.; Hemminger, J. C.; Finlayson-Pitts, B. J. Experimental and theoretical characterization of adsorbed water on self-assembled monolayers: Understanding the interaction of water with atmospherically relevant surfaces. J. Phys. Chem. A 2009, 113, 2060−2069. (34) Winter, N.; Vieceli, J.; Benjamin, I. Hydrogen-bond structure and dynamics at the interface between water and carboxylic acid functionalized self-assembled monolayers. J. Phys. Chem. B 2008, 112, 227−231. (35) Á lvarez, F.; Flores, E. A.; Castro, L. V.; Hernández, J. G.; López, A.; Vázquez, F. Dissipative particle dynamics (DPD) study of crude

percentages to tune surface characteristics. DPD simulations performed with the two sets of repulsive parameters led to similar contact angle values. A good agreement was observed between experimental contact angles and those from numerical simulations. DPD is an interesting numerical tool to study wetting properties of such kinds of systems. In this context the method proposed here can be generalized and applied to different coated surfaces or to study the surface adsorption phenomena of molecules in different applications.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS David Ramı ́rez-Gutieŕ rez gratefully acknowledges financial support from CONACYT and CampusFrance. The authors thank Pr. Hervé Toulhoat and Drs. Benjamin Herzhaft, Rafael Lugo, and Yannick Peysson for fruitful discussions.



REFERENCES

(1) Anastasiadis, S. H. Development of functional polymer surfaces with controlled wettability. Langmuir 2013, 29, 9277−9290. (2) Ikada, Y. Surface modification of polymers for medical applications. Biomaterials 1994, 15, 725−736. (3) Yilgor, I.; Bilgin, S.; Isik, M.; Yilgor, E. Tunable wetting of polymer surfaces. Langmuir 2012, 28, 14808−14814. (4) Ionov, L.; Houbenov, N.; Sidorenko, A.; Stamm, M.; Minko, S. Smart microfluidic channels. Adv. Funct. Mater. 2006, 16, 1153−1160. (5) Zhu, Y.; Fang, Q. Analytical detection techniques for droplet microfluidics-A review. Anal. Chim. Acta 2013, 787, 24−35. (6) Fadaei, H.; Scarff, B.; Sinton, D. Rapid microfluidics-based measurement of CO2 diffusivity in bitumen. Energy Fuels 2011, 25, 4829−4835. (7) Bowden, S. A.; Monaghan, P. B.; Wilson, R.; Parnell, J.; Cooper, J. M. The liquid-liquid diffusive extraction of hydrocarbons from a North Sea oil using a microfluidic format. Lab Chip 2006, 6, 740−743. (8) Bowden, S. A.; Wilson, R.; Parnell, J.; Cooper, J. M. Determination of the asphaltene and carboxylic acid content of a heavy oil using a microfluidic device. Lab Chip 2009, 9, 828−732. (9) Krebs, T.; Schroën, C. G. P. H.; Boom, R. M. Coalescence kinetics of oil-in-water emulsions studied with microfluidics. Fuel 2013, 106, 327−344. (10) Nilsson, M. A.; Kulkarni, R.; Gerberich, L.; Hammond, R.; Singh, R.; Baumhoff, E.; Rothstein, J. P. Effect of fluid rheology on enhanced oil recovery in a microfluidic sandstone device. J. NonNewtonian Fluid Mech. 2013, 202, 112−119. (11) Hemmilä, S.; Cauich-Rodriguez, J. V.; Kreutzer, J.; Kallio, P. Rapid, simple, and cost-effective treatments to achieve long-term hydrophilic PDMS surfaces. Appl. Surf. Sci. 2012, 258, 9864−9875. (12) Wang, L.; Sun, B.; Ziemer, K.S.; Barabino, G. A.; Carrier, R. L. Chemical and physical modifications to poly(dimethylsiloxane) surfaces affect adhesion of Caco-2 cells. J. Biomed. Mater. Res., Part A 2010, 93, 1260−1271. (13) Patrito, N.; McCague, C.; Chiang, S.; Norton, P. R.; Petersen, N. O. Photolithographically patterned surface modification of poly(dimethylsiloxane) via UV-initiated graft polymerization of acrylates. Langmuir 2013, 29, 9277−9290. (14) Li, J.; Wang, M.; Shen, Y. Chemical modification on top of nanotopography to enhance surface properties of PDMS. Surf. Coat. Technol. 2012, 206, 2161−2167. (15) Ginn, B. T.; Steinbock, O. Polymer surface modification using microwave-oven-generated plasma. Langmuir 2003, 19, 8117−8118. I

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir oil−water emulsions in the presence of a functionalized co-polymer. Energy Fuels 2011, 25, 562−567. (36) Liu, M. B.; Liu, G. R.; Zhou, L. W.; Chang, J. Z. Dissipative Particle Dynamics (DPD): An overview and recent developments. Arch. Comput. Methods Eng. 2014, DOI: 10.1007/s11831-014-9124-x. (37) Creton, B.; Nieto-Draghi, C.; Pannacci, N. Prediction of surfactants’ properties using multiscale molecular modeling tools: A review. Oil Gas Sci. Technol. 2012, 67, 969−982. (38) Chen, Z.; Cheng, X.; Cui, H.; Cheng, P.; Wang, H. Dissipative particle dynamics simulation of the phase behavior and microstructure of CTAB/octane/1-butanol/water microemulsion. Colloids Surf., A 2007, 301, 437−443. (39) Cupelli, C.; Henrich, B.; Glatzel, T.; Zengerle, R.; Moseler, M.; Santer, M. Dynamic capillary wetting studied with dissipative particle dynamics. New J. Phys. 2008, 10, 043009. (40) Liu, M.B.; Chang, J.Z.; Liu, H.T.; Su, T.X. Modeling of contact angles and wetting effects with particles methods. Int. J. Comput. Methods 2011, 8, 637−651. (41) Lin, T.-H.; Shih, W.-P.; Chen, C.-S.; Chiu, Y.-T. Simulation and analysis of interfacial wettability by dissipative particle dynamics. Proceedings of 1st IEEE International Conference on Nano/Micro Engineered and Molecular Systems, Zhuhai, China, Jan 18−21, 2006, 4135020, pp 571−574. (42) Paunov, V. N. Novel method for determining the three-phase contact angle of colloid particles adsorbed at air−water and oil−water interfaces. Langmuir 2003, 19, 7970−7976. (43) Jiang, S.; Granick, S. A simple method to produce trivalent colloidal particles. Langmuir 2009, 25, 8915−8918. (44) Willaime, H.; Pannacci, N.; Grasland-Mongrain, P.; Soares, E. M.; Benzaquen, M.; Tabeling, P. Generating double emulsions W/O/ W in a PDMS system by controlling locally the wetting properties of the channel. Proceedings of the 1st European Conference on Microfluidics − Microfluidics 2008, Bologna, Dec 10−12, 2008. (45) Schneider, M. H.; Tran, Y.; Tabeling, P. Benzophenone absorption and diffusion in poly(dimethylsiloxane) and its role in graft photo-polymerization for surface modification. Langmuir 2011, 27, 1232−1240. (46) Almutairi, Z.; Ren, C. L.; Simon, L. Evaluation of polydimethylsiloxane (PDMS) surface modification approaches for microfluidic applications. Colloids Surf., A 2012, 415, 406−412. (47) Hoogerbruge, P. J.; Koelman, J. M. V. A. Simulating microscopic hydrodynamic phenomena with dissipative particle dynamics. Europhys. Lett. 1992, 19, 155−160. (48) Vattulainen, I.; Karttunen, M.; Besold, G.; Polson, J. M. Integration schemes for dissipative particle dynamics simulations: From softly interacting systems towards hybrid models. J. Chem. Phys. 2002, 116, 3967−3979. (49) Groot, R. D.; Rabone, K. L. Mesoscopic simulation of cell membrane damage, morphology change and rupture by nonionic surfactants. Biophys. J. 2001, 81, 725−736. (50) Español, P. Hydrodynamics from dissipative particle dynamics. Phys. Rev. E 1995, 52, 1734−1742. (51) Zhang, S.-F.; Sun, L.-L.; Xu, J.-B.; Wu, H.; Wen, H. Aggregate structure in heavy crude oil: Using a dissipative particle dynamics based mesoscale platform. Energy Fuels 2010, 24, 4312−4326. (52) Deguillard, E.; Pannacci, N.; Creton, B.; Rousseau, B. Interfacial tension in oil−water−surfactant systems: On the role of intramolecular forces on interfacial tension values using DPD simulations. J. Chem. Phys. 2013, 138, 144102. (53) Groot, R. D.; Warren, P. B. Dissipative particle dynamics: Bridging the gap between atomistic and mesoscopic simulation. J. Chem. Phys. 1997, 107, 4423−4435. (54) Fan, C. F.; Olafson, B. D.; Blanco, M.; Hsu, S. L. Application of molecular simulation to derive phase diagrams of binary mixtures. Macromolecules 1992, 25, 3667−3676. (55) Materials Studio V 7.0; Accelrys Inc., San Diego, CA, 2013. (56) Maiti, A.; McGrother, S. Bead-bead interaction parameters in dissipative particle dynamics: Relation to bead-size, solubility parameter, and surface tension. J. Chem. Phys. 2004, 120, 1594−1601.

(57) Sun, H.; Ren, P.; Fried, J. R. The COMPASS forcefield: Parameterization and validation for polyphosphazenes. Comput. Theor. Polym. Sci. 1998, 8, 229−246. (58) Sun, H. COMPASS: An ab initio forcefield optimized for condensed-phase application-overview with details on alkane and benzene compounds. J. Phys. Chem. B 1998, 102, 7338−7364. (59) Chen, X. P.; Yuan, C. A.; Wong, C. K. L.; Zhang, G. Q. Validation of forcefields in predicting the physical and thermophysical properties of emeraldine base polyaniline. Mol. Simul. 2011, 37, 990− 996. (60) Baron, R.; Trzesniak, D.; de Vries, A. H.; Elsener, A.; Marrink, S. J.; van Gunsteren, W. F. Comparison of thermodynamic properties of coarse-grained and atomic-level simulation models. ChemPhysChem 2007, 8, 452−461. (61) Lee, J. N.; Park, C.; Whitesides, G. M. Solvent compatibility of poly(dimethylsiloxane)-based microfluidic devices. Anal. Chem. 2003, 75, 6544−6554. (62) Belmares, M.; Blanco, M.; Goddard, W. A., III; Ross, R. B.; Caldwell, G.; Chou, S.-H.; Pham, J.; Olofson, P. M.; Thomas, C. Hildebrand and Hansen Solubility parameters from molecular dynamics with applications to electronic nose polymer sensors. J. Comput. Chem. 2004, 25, 1814−1826.

J

DOI: 10.1021/la503694h Langmuir XXXX, XXX, XXX−XXX