Effect of Surface Tension, Viscosity, and Process Conditions on

Sep 5, 2013 - Mork Family Department of Chemical Engineering and Materials Science, ... (1-5) Monomer and initiator molecules are flown into a vacuum ...
2 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Effect of Surface Tension, Viscosity, and Process Conditions on Polymer Morphology Deposited at the Liquid−Vapor Interface Patrick D. Haller, Laura C. Bradley, and Malancha Gupta* Mork Family Department of Chemical Engineering and Materials Science, University of Southern California, Los Angeles, California, 90089, United States ABSTRACT: We have observed that the vapor-phase deposition of polymers onto liquid substrates can result in the formation of polymer films or particles at the liquid−vapor interface. In this study, we demonstrate the relationship between the polymer morphology at the liquid−vapor interface and the surface tension interaction between the liquid and polymer, the liquid viscosity, the deposition rate, and the deposition time. We show that the thermodynamically stable morphology is determined by the surface tension interaction between the liquid and the polymer. Stable polymer films form when it is energetically favorable for the polymer to spread over the surface of the liquid, whereas polymer particles form when it is energetically favorable for the polymer to aggregate. For systems that do not strongly favor spreading or aggregation, we observe that the initial morphology depends on the deposition rate. Particles form at low deposition rates, whereas unstable films form at high deposition rates. We also observe a transition from particle formation to unstable film formation when we increase the viscosity of the liquid or increase the deposition time. Our results provide a fundamental understanding about polymer growth at the liquid−vapor interface and can offer insight into the growth of other materials on liquid surfaces. The ability to systematically tune morphology can enable the production of particles for applications in photonics, electronics, and drug delivery and films for applications in sensing and separations.

■

INTRODUCTION The initiated chemical vapor deposition (iCVD) technique is a one-step, soventless process used to deposit thin functional polymer coatings.1−5 Monomer and initiator molecules are flown into a vacuum chamber where the initiator molecules are broken into radicals by a heated filament array. The monomer molecules and initiator radicals diffuse to the surface of a cooled substrate where polymerization occurs via a free-radical mechanism. The iCVD process is typically used to coat solid substrates,6−9 but we recently demonstrated the deposition of polymers onto liquids with low vapor pressures such as ionic liquids (ILs)10,11 and silicone oils.12 We observed the formation of both polymer films and particles at the liquid−vapor interface. Our previous studies focused on fabricating freestanding films12 and core−shell particles.11 In our current study, we systematically varied the surface tension interaction between the liquid and polymer, the liquid viscosity, and the process conditions in order to determine which parameters control polymer morphology at the liquid−vapor interface. Although we are the first group to study the deposition of polymers via iCVD onto liquid substrates, the deposition of parylene and inorganic materials onto liquid substrates has also been recently studied. Researchers have reported the formation of both films and particles, which is consistent with our observations. For example, Binh-Khiem et al. deposited transparent parylene films onto glycerin, liquid paraffin, and silicone oil. 13,14 Other studies have demonstrated the deposition of silver,15,16 gold,16 and copper17 particles onto silicone oils and gold particles onto ILs.18 Worden and co© 2013 American Chemical Society

workers found that depositing silver onto 1-ethyl-3-methylimidazolium ethylsulfate resulted in the formation of particles, whereas depositing chromium resulted in the formation of films because chromium has a higher nucleation density and deposition rate than silver.19 The results of our current study could help further explain the morphological differences observed in other deposition processes. In this study, we demonstrate that the thermodynamically stable morphology of polymers deposited onto the liquid− vapor interface via iCVD depends on whether it is energetically favorable for the polymer to spread over the liquid or aggregate. We found that unstable films can be formed by increasing the liquid viscosity, the deposition rate, or the deposition time. The insight gained from our study can allow for the production of nanoparticles for applications in photonics,20 electronics,21 and drug delivery22−24 as well as films for applications in sensing25,26 and separations.27

■

EXPERIMENTAL SECTION

Glycerol (EMD Chemicals), 1-ethyl-3-methylimidazolium tetrafluoroborate ([emim][BF4], 97%, Aldrich), 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF4], 97%, Aldrich), 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF6], 97%, Aldrich), squalene (98%, Sigma), 1-decyl-3-methylimidazolium tetrafluoroborate ([dmim][BF4], 96.5%, Aldrich), low viscosity silicone oil (Xiameter Received: July 4, 2013 Revised: August 18, 2013 Published: September 5, 2013 11640

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645

Langmuir

Article

were no polymer particles. A [bmim][BF4] droplet with a PnBA film was then separated and analyzed 24 h after deposition. The results showed the presence of particles with an average hydrodynamic radius of 139 ± 6 nm, which verifies that particles were formed when the film dewetted from the liquid surface after 24 h. A goniometer (Ramé-Hart Model 290-F1) was used to measure the advancing and equilibrium contact angles of the liquids on films of each polymer, the liquid−vapor surface tensions, and the polymer− vapor surface tensions. Each measurement was taken 7 times and averaged. The advancing contact angles were measured with an automated liquid dispensing system and the equilibrium contact angles were measured by dispensing 5 μL of liquid, allowing the droplets to equilibrate for 5 min, and then measuring the contact angles. The pendant drop method was used to measure liquid−vapor surface tensions.29 The acid−base method was used to calculate polymer− vapor surface tensions from the equilibrium contact angles of water, glycerol, and methylene iodide on films of each polymer.30 The errors in the measurements of the advancing contact angles, the polymer− vapor surface tensions, and the liquid−vapor surface tensions were estimated by averaging the standard deviations of each individual data point and were 2°, 1.5 mN/m, and 1.0 mN/m, respectively. The error in the calculation of the spreading coefficient was determined by propagating the errors31 through eq 1, which resulted in an estimated error of 3.5 mN/m. A quartz crystal microbalance with a 6 MHz gold-plated crystal was used to estimate the adsorption of 4VP onto the liquid−vapor interface and the absorption of 4VP into the 350 cP viscosity silicone oil. The measurements were performed at the same pressure, temperature, and flow rates as the polymer depositions; however, the filament was not heated in order to prevent polymerization. The monomer adsorption at the liquid−vapor interface was estimated by measuring the adsorption of 4VP onto the bare QCM crystal, while the absorption of 4VP into silicone oil was determined by measuring the total mass uptake of 4VP on a QCM crystal covered with a thin layer of silicone oil. The measured mass uptake of 4VP onto the bare crystal and the crystal coated with silicone oil were similar indicating that 4VP adsorbed onto the silicone oil surface but did not absorb into the silicone oil.

PMX-200 350 cP, Aldrich), high viscosity silicone oil (100,000 cP, Aldrich), 1H,1H,2H,2H-perfluorodecyl acrylate (PFDA, 97%, Synquest), pentafluorophenyl methacrylate (PFM, 97%, Synquest), nbutyl acrylate (nBA, 99%, Aldrich), n-butyl methacrylate (nBMA, 99%, Aldrich), 2-hydroxyethyl methacrylate (HEMA, 97%, Aldrich), 4vinylpyridine (4VP, 95%, Aldrich) and tert-butyl peroxide (98%, Aldrich) were used as received without further purification. Liquid droplets (5 μL) were dispensed from a micropipet onto silicon wafers which were placed in a custom designed reaction chamber (250 mm diameter, 48 mm height, GVD Corporation). For all depositions, the initiator was kept at room temperature and flown into the reactor at a rate of 1.0 sccm. Inside the reactor, a nichrome filament array (80% Ni, 20% Cr, Omega Engineering) was held 32 mm above that stage and resistively heated to 200 °C. The temperature of the stage was controlled using a recirculating backside chiller. Specific reaction conditions for each of the polymers (poly(1H,1H,2H,2Hperfluorodecyl acrylate) (PPFDA), poly(pentafluorophenyl methacrylate) (PPFM), poly(n-butyl acrylate) (PnBA), poly(n-butyl methacrylate) (PnBMA), poly(2-hydroxyethyl methacrylate) (PHEMA), and poly(4-vinylpyridine) (P4VP)) can be found in Table 1. The deposition rate was determined by measuring the film thickness on a reference silicon wafer using a profilometer (Ambios Technology XP2).

Table 1. Summary of the Deposition Conditions Used in This Study monomer

monomer jar temp. (°C)

monomer flow rate (sccm)

stage temp. (°C)

reactor pressure (mTorr)

deposition rate (nm/min)

PFDA

50

1.2

35

PFM nBA

25 25

1.5 4.2

20 10

55 80 100 120 100 300 450 600 750 900 200 400 600 800 100 210 520 830 1000

3 10 27 37 10 10 23 29 43 51 10 37 50 80 10 6 10 33 44

nBMA

25

6.5

10

HEMA 4VP

50 25

0.8 4.1

35 25

■

RESULTS AND DISCUSSION In order to study the effect of surface tension on the initial polymer morphology deposited at the liquid−vapor interface, we deposited a range of polymers onto a variety of liquid substrates at a rate of 10 nm/min as measured on a reference silicon wafer. We conducted short depositions of 5 min in order to minimize the effect of polymer accumulation on the liquid surface tension and viscosity. The surface tension interaction between the polymer and the liquid substrate was quantified by the spreading coefficient, S, which is a measure of the free energy required for the polymer to spread over the surface of the liquid.29,32 A positive spreading coefficient indicates that it is energetically favorable for the polymer to spread over the surface of the liquid, whereas a negative spreading coefficient indicates that it is energetically favorable for the polymer to reduce the area of contact with the liquid surface. The spreading coefficient can be written in terms of the liquid− vapor surface tension (γLV), the polymer−vapor surface tension (γPV), and the advancing contact angle of the liquid on the polymer (θ):33

The polymer deposited onto the liquid droplets was imaged using a stereo microscope at 10× total magnification. The P4VP particles shown in Figure 1 were separated from the silicone oil surface using a 0.01% v/v solution of Triton-X surfactant in water. After separation, 50 μL of the solution was drop-cast onto a clean silicon wafer and allowed to dry under ambient conditions. The sample was sputter coated with gold and imaged using a JSM-7001 low-vacuum scanning electron microscope. Dynamic light scattering (Wyatt DynaPro Titan) was used to confirm the particle size in a 0.01% v/v Triton-X solution. Measurements of 5 samples showed that the particles had an average hydrodynamic radius of 126 ± 4 nm. Dynamic light scattering was also used to verify that the PnBA film that dewetted from the [bmim][BF4] surface after 24 h formed particles. As a control sample, a [bmim][BF4] droplet with a PnBA film was centrifuged immediately after deposition, the IL was then removed with a 25% v/v solution of methanol in water, and the polymer was dispersed in a 0.01% v/v Triton-X in water solution for analysis. The results showed only the presence of Triton-X micelles of 0.7 nm radius,28 indicating that there

S = γLV*(1 + cos θ ) − 2γPV

(1)

In order to calculate the spreading coefficient for each polymer−liquid system, we used a goniometer to measure the parameters in eq 1 (Table 2). We found that polymer films formed when it was energetically favorable for the polymer to spread over the liquid surface (positive spreading coefficients), 11641

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645

Langmuir

Article

liquid, leading to different polymer morphologies. For example, PnBA (γPV = 35.1 mN/m) has a positive spreading coefficient on the high surface tension liquid glycerol (S = 4.3 mN/m) and formed a film, whereas PnBA has a negative spreading coefficient on the low surface tension liquid silicone oil (S = −32.8 mN/m) and formed particles. The measurement error in the spreading coefficients reported in Table 2 is ∼3.5 mN/m which likely explains why some systems with slightly negative spreading coefficients form films (i.e., PnBMA on [emim][BF4] and PPFM on [bmim][PF6]) and some systems with slightly positive spreading coefficients form particles (i.e., PnBA on [bmim][PF6] and PPFM on squalene). It is important to note that we previously showed that monomer molecules can absorb into the liquid in some polymer−liquid systems.10 For example, the extent of monomer absorption into ionic liquids has been shown to depend on the alkyl chain length of the cation,34 the nucleophilicity of the anion,35 and the polarity of the monomer.36 The absorption of monomer into the liquid can alter the liquid−vapor surface tension and the advancing contact angle, leading to changes in the spreading coefficient. We ignored these effects in our calculations of the spreading coefficients because we always observed films for the systems with positive spreading coefficients that absorb monomer (i.e., nBA in [emim][BF4]), indicating that the spreading coefficient remained positive, and we always observed particles for the systems with negative spreading coefficients that absorb monomer (i.e., HEMA in [emim][BF4]), indicating that the spreading coefficient remained negative. The spreading coefficient determines the thermodynamically preferred morphology of the polymer but does not account for the kinetic limitations of the deposition process; therefore, we tested whether the initial morphology could be tuned by varying the polymer deposition rate (Dr). To study the effect of the deposition rate, we varied the monomer concentration at the liquid−vapor interface by varying the reactor pressure37 and continued to use short deposition times of 5 min. For a wide range of deposition rates, we found that systems with highly positive spreading coefficients always formed films and systems with highly negative spreading coefficients always formed particles. For example, the deposition of PPFDA onto [emim][BF4] (S = 11.8 mN/m) resulted in films for Dr = 3−37 nm/min, whereas the deposition of P4VP onto silicone oil (S = −71.6 mN/m) resulted in particles for Dr = 6−44 nm/ min. For systems with moderate spreading coefficients that do not strongly favor spreading or aggregation (S ≈ 0), we found that particles formed at low deposition rates and films formed at high deposition rates. For example, the deposition of PnBA onto [bmim][BF4] (S = −4.2 mN/m) resulted in particles for Dr = 10, 23, and 29 nm/min and clear films for Dr = 43 and 51 nm/min. We observed a similar morphological dependence on deposition rate for PnBA on [bmim][PF6] (S = 0.8 mN/m) and PnBMA on [bmim][PF6] (S = −0.4 mN/m) (Table 3). We also conducted a 1 min PnBA deposition onto [bmim][BF4] at Dr = 51 nm/min to match the amount of polymer deposited in 5 min at Dr = 10 nm/min and we observed a film, which confirms that the film was formed because of an increased rate of deposition rather than an increased amount of total polymer. For the systems with moderate spreading coefficients, we hypothesize that particles form at low deposition rates because the polymer chains are able to diffuse across the liquid surface, whereas at high deposition rates the polymer chains have a higher molecular weight38 causing the chains to diffuse more slowly and overlap

Table 2. Summary of the Spreading Coefficients in mN/m (bold), Liquid−Vapor Surface Tensions (γLV), Polymer− Vapor Surface Tensions (γPV), and Advancing Contact Angles for All Tested Polymer−Liquid Systemsa

Unshaded entries indicate systems that formed films and shaded entries indicate systems that formed particles for 5 min depositions at a rate of 10 nm/min. a

whereas polymer particles formed when it was energetically favorable for the polymer to aggregate (negative spreading coefficients). For example, PPFDA on [emim][BF4] has a positive spreading coefficient (S = 11.8 mN/m) and formed a film (Figure 1A), whereas P4VP on silicone oil has a negative

Figure 1. Images showing (A) a PPFDA polymer film on [emim][BF4] (S = 11.8 mN/m) and (B) P4VP polymer particles on silicone oil (S = −71.6 mN/m). (C,D) Scanning electron microscope images of P4VP particles which were removed from the silicone oil and drop-cast onto a silicon wafer.

spreading coefficient (S = −71.6 mN/m) and formed particles (Figure 1B). The P4VP particles were recovered from the silicone oil surface and analyzed using scanning electron microscopy (Figure 1C,D) and dynamic light scattering to confirm that discrete nanoparticles were formed. It is important to note that the low surface tension polymer PPFDA (γPV = 13.6 mN/m) has a positive spreading coefficient on each of the liquids tested and formed films in all of these cases. These films remained clear and unchanged for more than one month at room temperature, indicating that the films were thermodynamically stable. In contrast, the high surface tension polymers PHEMA (γPV = 48.5 mN/m) and P4VP (γPV = 57.5 mN/m) have negative spreading coefficients on all of the liquids tested and formed particles in all of these cases. For polymers with intermediate surface tensions, the sign of the spreading coefficient varied depending on the surface tension of the 11642

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645

Langmuir

Article

formation for systems that thermodynamically favor particles. We hypothesize that increasing the liquid viscosity from 350 to 100,000 cP leads to a decrease in the polymer diffusion coefficient by several orders on magnitude enabling the polymer chains to overlap on the liquid surface before they can aggregate. The films formed on the high viscosity silicone oil were not thermodynamically stable due to the negative spreading coefficient and we observed that the films dewetted within 5 days. Our studies mentioned above were conducted using short 5 min depositions in order to minimize the effects of polymer accumulation at the liquid−vapor interface. In order to test the effect of polymer accumulation, we deposited P4VP at Dr = 10 nm/min for deposition times ranging from 5 to 180 min. We found that particles were initially formed on all of the liquids tested, but longer deposition times resulted in films (Table 4).

Table 3. Effect of Deposition Rate on the Initial Polymer Morphology for Polymer−Liquid Systems with Moderate Spreading Coefficients (S ≈ 0) liquid

S (mN/m)

deposition rate (nm/min)

morphology

PnBA

[bmim][BF4]

−4.2

PnBMA

[bmim][PF6]

−0.4

PnBA

[bmim][PF6]

0.8

10 23 29 43 51 10 37 50 80 10 23 29 43 51

Particles Particles Particles Films Films Particles Particles Films Films Particles Films Films Films Films

polymer

Table 4. Deposition of P4VP onto a Range of Liquids Initially Results in the Formation of Particles (Shaded) and Increasing the Deposition Time Leads to the Formation of Films (Unshaded)

more rapidly,39 leading to the formation of a film. Unlike the thermodynamically stable films formed in systems with highly positive spreading coefficients (i.e., PPFDA films), the films formed at high deposition rates for systems with moderate spreading coefficients were not thermodynamically stable and eventually dewetted from the liquid surface. For example, the clear PnBA films deposited at Dr = 43 nm/min onto [bmim][BF4] (S = −4.2 mN/m) dewetted and formed particles after 24 h, which was verified using dynamic light scattering. The Rouse model for diffusion can be applied to polymers at liquid−vapor interfaces40 in which the diffusion coefficient is inversely proportional to the liquid viscosity.41 We therefore hypothesized that we could favor the formation of unstable films instead of particles for systems with negative spreading coefficients by using higher viscosity liquids. To test the effect of liquid viscosity on the polymer morphology, we compared the deposition of P4VP onto 350 cP and 100,000 cP silicone oils at 10 nm/min for 5 min. We chose the model system of 4VP and silicone oil because our quartz crystal microbalance measurements indicate that 4VP is insoluble in silicone oil, and therefore the spreading coefficients and liquid viscosities will not change due to monomer absorption. P4VP has similar spreading coefficients on both 350 cP (S = −71.6) and 100,000 cP silicone oils (S = −78.8 mN/m) due to similar advancing contact angles (25° and 32°, respectively) and similar liquid− vapor surface tensions (22.8 and 19.6 mN/m, respectively). Despite similar spreading coefficients, we found that particles formed on the 350 cP silicone oil (Figure 2A), whereas clear films formed on the 100,000 cP silicone oil (Figure 2B) demonstrating that very high viscosity liquids can enable film

For example, a 5 min deposition of P4VP onto [emim][BF4] resulted in the formation of particles (Figure 3A), while a 60

Figure 3. Top-down images show that the deposition of P4VP onto [emim][BF4] results in (A) particles after 5 min and (B) a film after 60 min.

min deposition resulted in a film (Figure 3B). The transition from particle formation to film formation at long deposition times could result from either the coalescence of particles to form a continuous film42,43 or an increase in the viscosity at the liquid surface due to the accumulation of particles44 leading to film formation similar to the formation of the P4VP film on the high viscosity silicone oil. We also observed that the time it took for a film to form decreased with increasing equilibrium contact angle (θe) of the liquid on the polymer. For example, [emim][BF4] has a high contact angle on P4VP (θe = 56°) and less than 15 min of deposition was required to form a film, whereas [dmim][BF4] has a low contact angle on P4VP (θe = 11°) and more than 120 min of deposition was required to form a film. Increasing the equilibrium contact angle between 0° and 90°, which reflects the range of contact angles for polymer−liquid systems that form particles, results in an increase in the distance that the particles protrude from the surface of the liquid45 which may either provide more exposed

Figure 2. Deposition of P4VP at a rate of 10 nm/min for 5 min results in (A) particles on 350 cP silicone oil and (B) a clear film on 100,000 cP silicone oil. 11643

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645

Langmuir

Article

(8) O’Shaughnessy, W. S.; Murthy, S. K.; Edell, D. J.; Gleason, K. K. Stable Biopassive Insulation Synthesized by Initiated Chemical Vapor Deposition of Poly(1,3,5-trivinyltrimethylcyclotrisiloxane). Biomacromolecules 2007, 8, 2564−2570. (9) Gupta, M.; Kapur, V.; Pinkerton, N. M.; Gleason, K. K. Initiated Chemical Vapor Deposition (iCVD) of Conformal Polymeric Nanocoatings for the Surface Modification of High-Aspect-Ratio Pores. Chem. Mater. 2008, 20, 1646−1651. (10) Haller, P. D.; Frank-Finney, R. J.; Gupta, M. Vapor-Phase Free Radical Polymerization in the Presence of an Ionic Liquid. Macromolecules 2011, 44, 2653−2659. (11) Bradley, L. C.; Gupta, M. Encapsulation of Ionic Liquids within Polymer Shells via Vapor Phase Deposition. Langmuir 2012, 28, 10276−10280. (12) Frank-Finney, R. J.; Haller, P. D.; Gupta, M. Ultrathin FreeStanding Polymer Films Deposited onto Patterned Ionic Liquids and Silicone Oil. Macromolecules 2012, 45, 165−170. (13) Binh-Khiem, N.; Matsumoto, K.; Shimoyama, I. Polymer thin film deposited on liquid for varifocal encapsulated liquid lenses. Appl. Phys. Lett. 2008, 93, 124101. (14) Binh-Khiem, N.; Matsumoto, K.; Shimoyama, I. Tensile Stress of Parylene Deposited on Liquid. Langmuir 2010, 26, 18771−18775. (15) Ye, G.-X.; Michely, T.; Weidenhof, V.; Friedrich, I.; Wuttig, M. Nucleation, Growth, and Aggregation of Ag Clusters on Liquid Surfaces. Phys. Rev. Lett. 1998, 81, 622−625. (16) Xie, J.-P.; Yu, W.-Y.; Zhang, S.-L.; Chen, M.-G.; Ye, G.-X. AFM study on microstructures of metal films deposited on liquid substrates. Phys. Lett. A 2007, 371, 160−164. (17) Chen, M.-G.; Yu, S.-J.; Feng, Y.-X.; Jiao, Z.-W.; Yu, M.-Z.; Yang, B. Microstructure and growth mechanism of Cu ramified aggregates on silicone oil surfaces. Thin Solid Films 2010, 518, 2674−2677. (18) Torimoto, T.; Okazaki, K.; Kiyama, T.; Hirahara, K.; Tanaka, N.; Kuwabata, S. Sputter deposition onto ionic liquids: Simple and clean synthesis of highly dispersed ultrafine metal nanoparticles. Appl. Phys. Lett. 2006, 89, 243117. (19) Borra, E. F.; Seddiki, O.; Angel, R.; Eisenstein, D.; Hickson, P.; Seddon, K. R.; Worden, S. P. Deposition of metal films on an ionic liquid as a basis for a lunar telescope. Nature 2007, 447, 979−981. (20) Fudouzi, H.; Xia, Y. Photonic Papers and Inks: Color Writing with Colorless Materials. Adv. Mater. 2003, 15, 892−896. (21) Ling, Q.-D.; Liaw, D.-J.; Zhu, C.; Chan, D. S.-H.; Kang, E.-T.; Neoh, K.-G. Polymer electronic memories: Materials, devices and mechanisms. Prog. Polym. Sci. 2008, 33, 917−978. (22) Chouhan, R.; Bajpai, A. K. An in vitro release study of 5-fluorouracil (5-FU) from swellable poly-(2-hydroxyethyl methacrylate) (PHEMA) nanoparticles. J. Mater. Sci.: Mater. Med. 2009, 20, 1103− 1114. (23) Chuang, C.-Y.; Don, T.-M.; Chiu, W.-Y. Synthesis and properties of chitosan-based thermo- and pH-responsive nanoparticles and application in drug release. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 2798−2810. (24) Zhang, Z.; Grijpma, D. W.; Feijen, J. Poly(trimethylene carbonate) and monomethoxy poly(ethylene glycol)-block-poly(trimethylene carbonate) nanoparticles for the controlled release of dexamethasone. J. Controlled Release 2006, 111, 263−270. (25) Jiang, C.; Markutsya, S.; Pikus, Y.; Tsukruk, V. V. Freely suspended nanocomposite membranes as highly sensitive sensors. Nat. Mater. 2004, 3, 721−728. (26) Zhai, L.; Nolte, A. J.; Cohen, R. E.; Rubner, M. F. pH-Gated Porosity Transitions of Polyelectrolyte Multilayers in Confined Geometries and Their Application as Tunable Bragg Reflectors. Macromolecules 2004, 37, 6113−6123. (27) Zimnitsky, D.; Shevchenko, V. V.; Tsukruk, V. V. Perforated, Freely Suspended Layer-by-Layer Nanoscale Membranes. Langmuir 2008, 24, 5996−6006. (28) Robson, R. J.; Dennis, E. A. The Size, Shape, and Hydration of Nonionic Surfactant Micelles. Triton X-100. J. Phys. Chem. 1977, 81, 1075−1078.

surface area for particles to coalesce or result in a larger increase in surface viscosity,46 leading to the formation of films at shorter deposition times. The P4VP films that formed at long deposition times were unstable due to the negative spreading coefficients and dewetted within 24 h.

■

CONCLUSIONS We have demonstrated that the thermodynamically stable morphology of the polymer deposited onto liquid substrates depends on the spreading coefficient of the polymer on the liquid. For systems with positive spreading coefficients, it is energetically favorable for the polymer to spread across the liquid surface resulting in a film, whereas for systems with negative spreading coefficients it is energetically favorable for the polymer to aggregate into particles in order to reduce the area of contact with the liquid surface. For systems that do not strongly favor spreading or aggregation, the initial morphology depends on the deposition rate, with particles forming at low deposition rates and unstable films forming at high deposition rates. We also found that unstable films can be formed for systems with negative spreading coefficients by increasing the viscosity of the liquid or by increasing the deposition time. This study has enhanced the understanding of dynamic polymer− liquid interfacial phenomena and the fundamental insight gained may be applicable to other vapor-phase deposition processes such as parylene deposition and metal sputtering.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge USC start-up funds. P.D.H. is supported by a USC NSF GK-12 graduate fellowship (DGE-1045595). L.C.B. is supported by a fellowship from the Chevron Corporation (USC−CVX UPP).

■

REFERENCES

(1) Gupta, M.; Gleason, K. K. Initiated Chemical Vapor Deposition of Poly(1H,1H,2H,2H-perfluorodecyl Acrylate) Thin Films. Langmuir 2006, 22, 10047−10052. (2) Chan, K.; Gleason, K. K. Initiated Chemical Vapor Deposition of Linear and Cross-linked Poly(2-hydroxyethyl methacrylate) for Use as Thin-Film Hydrogels. Langmuir 2005, 21, 8930−8939. (3) Haller, P. D.; Flowers, C. A.; Gupta, M. Three-dimensional patterning of porous materials using vapor phase polymerization. Soft Matter 2011, 7, 2428−2432. (4) Tenhaeff, W. E.; McIntosh, L. D.; Gleason, K. K. Synthesis of Poly(4-vinylpyridine) Thin Films by Initiated Chemical Vapor Deposition (iCVD) for Selective Nanotrench-Based Sensing of Nitroaromatics. Adv. Funct. Mater. 2010, 20, 1144−1151. (5) Alf, M. E.; Godfrin, P. D.; Hatton, T. A.; Gleason, K. K. Sharp Hydrophilicity Switching and Conformality on Nanostructured Surfaces Prepared via Initiated Chemical Vapor Deposition (iCVD) of a Novel Thermally Responsive Copolymer. Macromol. Rapid Commun. 2010, 31, 2166−2172. (6) Ma, M.; Mao, Y.; Gupta, M.; Gleason, K. K.; Rutledge, G. C. Superhydrophobic Fabrics Produced by Electrospinning and Chemical Vapor Deposition. Macromolecules 2005, 38, 9742−9748. (7) Lau, K. K. S.; Gleason, K. K. Particle Surface Design using an AllDry Encapsulation Method. Adv. Mater. 2006, 18, 1972−1977. 11644

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645

Langmuir

Article

(29) Adamson, A. W.; Gast, A. P. Physical Chemistry of Surfaces, 6th ed.; John Wiley & Sons, Inc.: New York, NY, 1997. (30) van Oss, C. J.; Good, R. J.; Chaudhury, M. K. Additive and Nonadditive Surface Tension Components and the Interpretation of Contact Angles. Langmuir 1988, 4, 884−891. (31) Taylor, J. R. An Introduction to Error Analysis: The Study of Uncertainties in Physical Measurements, 1st ed.; University Science Books: Mill Valley, CA, 1982. (32) Wyart, F. B.; Martin, P.; Redon, C. Liquid/Liquid Dewetting. Langmuir 1993, 9, 3682−3690. (33) Fox, H. W.; Zisman, W. A. The spreading of liquids on lowenergy surfaces. II. Modified tetrafluoroethylene polymers. J. Colloid Sci. 1952, 107, 109−121. (34) Muldoon, M. J.; Aki, S. N. V. K.; Anderson, J. L.; Dixon, J. K.; Brennecke, J. F. Improving Carbon Dioxide Solubility in Ionic Liquids. J. Phys. Chem. B 2007, 111, 9001−9009. (35) Muldoon, M. J.; Gordon, C. M.; Dunkin, I. R. Investigations of Solvent-Solute Interactions in Room-Temperature Ionic Liquids Using Solvatochromic Dyes. J. Chem. Soc., Perkins Trans. 2001, 2, 433−435. (36) Anthony, J. L.; Anderson, J. L.; Maginn, E. J.; Brennecke, J. F. Anion Effects on Gas Solubility in Ionic Liquids. J. Phys. Chem. B 2005, 109, 6366−6374. (37) Lau, K. K. S.; Gleason, K. K. Initiated Chemical Vapor Deposition (iCVD) of Poly(alkyl acrylates): An Experimental Study. Macromolecules 2006, 39, 3688−3694. (38) Lau, K. K. S.; Gleason, K. K. Initiated Chemical Vapor Deposition (iCVD) of Poly(alkyl acrylates): A Kinetic Model. Macromolecules 2006, 39, 3695−3703. (39) de Gennes, P.-G. Scaling Concepts in Polymer Physics, 1st ed.; Cornell University: Ithaca, NY, 1979. (40) Monroy, F.; Hilles, H. M.; Ortega, F.; Rubio, R. G. Relaxation Dynamics of Langmuir Polymer Films: A Power-Law Analysis. Phys. Rev. Lett. 2003, 91, 268302. (41) Lee, J. C.-M.; Santore, M.; Bates, F. S.; Discher, D. E. From Membranes to Melts, Rouse to Reptation: Diffusion in Polymersome versus Lipid Bilayers. Macromolecules 2002, 35, 323−326. (42) Floro, J. A.; Hearne, S. J.; Hunter, J. A.; Kotula, P.; Chason, E.; Seel, S. C.; Thompson, C. V. The dynamic competition between stress generation and relaxation mechanisms during coalescence of VolmerWeber thin films. J. Appl. Phys. 2001, 89, 4886−4897. (43) Craven, M. D.; Wu, F.; Chakraborty, A.; Imer, B.; Mishra, U. K.; DenBaars, S. P.; Speck, J. S. Microstructural evolution of a-plane GaN grown on a-plane SiC by metalorganic chemical vapor deposition. Appl. Phys. Lett. 2004, 84, 1281−1283. (44) Tambe, D. E.; Sharma, M. M. Factors Controlling the Stability of Colloid-Stabilized Emulsions. J. Colloid Interface Sci. 1994, 162, 1− 10. (45) McHale, G.; Newton, M. I. Liquid marbles: principles and applications. Soft Matter 2011, 7, 5473−5481. (46) Safouane, M.; Langevin, D.; Binks, B. P. Effect of Particle Hydrophobicity on the Properties of Silica Particle Layers at the Air− Water Interface. Langmuir 2007, 23, 11546−11553.

11645

dx.doi.org/10.1021/la402538e | Langmuir 2013, 29, 11640−11645