Atmospheric Pressure Plasma-Initiated Chemical Vapor Deposition

May 31, 2017 - (AP-PiCVD) of Poly(alkyl acrylates): An Experimental Study. François Loyer, Gilles Frache, Patrick Choquet, and Nicolas D. Boscher*...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/Macromolecules

Atmospheric Pressure Plasma-Initiated Chemical Vapor Deposition (AP-PiCVD) of Poly(alkyl acrylates): An Experimental Study François Loyer, Gilles Frache, Patrick Choquet, and Nicolas D. Boscher* Department of Materials Research and Technology, Luxembourg Institute of Science and Technology, L-4422 Belvaux, Luxembourg S Supporting Information *

ABSTRACT: A novel atmospheric pressure plasma-initiated chemical vapor deposition (AP-PiCVD) approach toward the growth of conventional polymer layers is characterized and interpreted. A set of three methacrylate monomers (methyl, butyl, and glycidyl methacrylate) were investigated using ultrashort plasma discharges (ca. 100 ns) pulsed at various frequencies, covering a range of duty cycle from 0.1% to 0.000 316%. An unprecedented weight-average molar mass of 94 000 g mol−1 coupled to an outstanding thin film conformality and an excellent chemical functionalities retention was achieved for the best deposition conditions. Insights into the growth mechanisms in AP-PiCVD and their dependence on the monomer’s intrinsic properties are provided.



deposition of polymer thin films from a dry method such as chemical vapor deposition (CVD). CVD processes are especially investigated because of their flexibility and proven advantages (e.g., solvent-free and one-step methods).3,6,9−11 More particularly, initiated CVD (iCVD) or oxidative CVD (oCVD) processes can yield fast, well-controlled, and highly reproducible formation12 of conformal polymer and copolymer layers.9,10,12,13 Although, as a result of the process itself, the available CVD methods toward the simultaneous synthesis and deposition of polymer layers are operating under vacuum, which remains an important drawback for upscaling to industrial level. Tremendous efforts have been made in the field of plasmaenhanced CVD (PECVD) toward the formation of organic coatings with a polymer-like chemistry.14,15 Among the advantages of PECVD processes, their ability to work under atmospheric-pressure and room temperature conditions16 is particularly noteworthy, especially when using an atmosphericpressure dielectric barrier discharge (AP-DBD) setup.17 Other significant assets of PECVD processes include their ability to form organic coatings in the absence of polymerizable bonds,18 the ability to mix monomers with various chemical functionalities or reactivities,19,20 and the potentiality to tune the surface morphology.21,22 However, plasmas, which are composed of a wide variety of reactive species including highly energetic elements, induce nonspecific reactions. As a consequence, the chemical structure of the monomers is only partially retained, and the formed materials often present a high degree of crosslinking, a much lower concentration of functional groups, and properties that can strongly differ from the counterpart

INTRODUCTION Polymers exhibit a large array of functional properties due to their wide variety of chemical groups1 and their micro- and nanostructural layout.2 Tuning of both those properties and chemical composition (including copolymers) enable the formation of an unlimited amount of compounds differing in chemical behavior, physical properties, responsivity, and physical state.3,4 Therefore, the means to selectively control their synthesis became one of the main challenges for researchers and a vital part of the world industry. In addition, the deposition of polymer thin films has also drawn a lot of interest due to their ability to grant their physical and chemical properties to a functionalized substrate,3−6 while being economical due to their small thicknesses. Consequently, lots of methods of deposition have been developed, each with their advantages and drawbacks, compromising between the cost and the efficiency. The cheapest, easiest to upscale at an industrial level and most widely adopted processes are based on the transfer of presynthesized polymer from a solution to a substrate.7,8 Several methods, such as dip-coating and spin-coating, have been thoroughly investigated through the years and optimized to have a better control of the thin film properties. In a general manner, the substrate is either immersed in or recovered by the polymer solution.7,8 The thickness is then controlled by the solvent distribution method7 (rising pace, spinning speed, flow velocity, etc.) and its evaporation time. Unfortunately, those methods carry unavoidable drawbacks such as the inability or difficulty to form conformal films, the lack of control of low thicknesses, the limitation of the polymer solubility, the necessity of post-treatment, and most importantly the need to dispose of large quantities of organic solvents. For ecological and practical arguments, there are undeniable motivations to move toward the simultaneous synthesis and © XXXX American Chemical Society

Received: March 6, 2017 Revised: May 22, 2017

A

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

another. Investigation and cross-comparison of the thin films’ deposition rate, morphology, chemistry, and polymeric growth highlighted not only a pattern of effects resulting from the plasma but also trends coherent with the monomers different kinetic properties. Thus, we open a discussion on the various mechanisms and features affecting the simultaneous polymerization and thin film formation in AP-PiCVD using indirect evidence, while highlighting its strong potential for tuning the polymers properties only using the discharge frequency. The defined guidelines are further implemented to produced polymer layers with weight-average molar masses as high as 94 000 g mol−1, which represents an unprecedented value for both plasma-enhanced and atmospheric-pressure CVD methods.

compounds synthesized by conventional methods.23 Therefore, the produced thin films cannot be considered as polymers and are often called “plasma polymers”.24,25 With the perspective of minimizing the negative impact of plasmas and form plasma-polymer thin films closer to conventional polymers, numerous works have investigated the use of pulsed plasma discharges at both low and atmospheric pressure.20,26−30 Klages et al. have notably shown, using a pulsed AP-DBD, that the free-radical polymerization and deposition of vinyl monomers can be performed in the gas phase under ambient conditions.26 Noticeable improvements of the functionalities retention were reported for plasma off-times (toff) several times longer than the plasma on-time (ton), i.e., for duty cycles (DC = ton/(ton + toff)) ranging from 0.006% to 33%.26 Nevertheless, if the investigated toff, in the range of several tens of milliseconds, was shown consistent with the lifetime of the free-radical propagation reaction, the influence of the plasma (ton = 1 ms) on the thin film chemistry remained a drawback for more advanced applications. Very recent studies have further improved the principle using ultrashort square pulse generator.31,32 Such an approach allowed a momentous reduction of the plasma on-time to several tens of nanoseconds, 4 orders of magnitude shorter than the sinusoidal signals traditionally employed in AP-DBD which do not allow ton lower than several hundreds of microseconds. The combination of extremely short plasma discharges (ca. 100 ns per pulse) and toff up to hundreds of milliseconds, corresponding to the freeradical polymerization lifetime, strongly promotes the conventional polymerization pathway while statically lowering the plasma-induced fragmentation and recombination.31 Indeed, during toff most of the reactive species created under plasma quickly lose their excited state, leaving only the long lifetime species, such as the growing polymer chains. The approach, called plasma-initiated chemical vapor deposition (PiCVD), led to the formation of polymer layers with a unprecedented degree of polymerization from plasma thin films, such as highlighted by mass spectrometry and size exclusion chromatography.31,32 To gain a deeper understanding of the AP-PiCVD method and evaluate its potential range of applications, we investigated a set of three methacrylate monomers, chosen according to their kinetic parameters of adsorption (Psat) and polymerization (kp). By keeping the plasma on-time constant and changing the discharge frequencies from 31.6 Hz (i.e., every 31.6 ms) up to 10 000 Hz (i.e., every 100 μs), we altered the deposition conditions and growth mechanisms from one extreme to



EXPERIMENTAL SECTION

Materials and PiCVD. The thin films presented in this work were deposited using an atmospheric-pressure dielectric barrier discharge setup as previously described33 (Figure 1a). The discharge gap was maintained to 1 mm. The plasma discharges were ignited by 15 μs square pulses of 6.5 kV, produced by an AHTPB10F generator from EFFITECH (Gif-sur-Yvette, France), allowing the generation of perfectly reproducible ultrashort plasma discharge (Figure 1b and Figure S1). For every experiment, all generator’s input parameters were kept identical except for the pulse repetition frequency varying from 31.6 up to 10 000 Hz (Table S1). A set of three different methacrylate monomers were chosen for a comparative study of their polymerization properties in AP-PiCVD, specifically because of their kinetic rate constantskp and ktand the vapor saturation pressure Psat. The deposition conditions are summarized in Table S1. The methacrylate monomersmethyl methacrylate (MMA, Sigma-Aldrich, 99%), butyl methacrylate (BMA, Sigma-Aldrich, 99%) and glycidyl methacrylate (GMA, Sigma-Aldrich, 97%)were used without any additional treatment and were directly supplied to the deposition area using a bubbler setup and argon (Air Liquid, 99.999%) as a carrier gas. The carrier flow was adapted for each monomer according to their vapor pressure in order to deliver the same amount of molecules per second to the deposition area. The total gas flow was maintained constant to 20 L min−1 using another Ar source. To avoid O2 and N2 contamination, argon fluxes were added in both sides of the electrode in order to remove as much atmosphere as possible. All depositions were carried on polished 4 in. silicon wafers (Siltronix, Archamps, France); prior to each experiment they were cleaned using a 95%/5% argon/oxygen plasma for 40 s. Polymer references were bought from chemical distributors: Sigma-Aldrich for both poly(methyl methacrylate) (pMMA) and poly(glycidyl methacrylate) (pGMA) and Polymer Source for poly(butyl methacrylate) (pBMA). Thin Film Characterizations. SEM imaging and thickness measurements were carried out on a Hitachi SU-70 FE-SEM (Tokyo, Japan). In order to avoid distortions due to charge effect, the samples

Figure 1. (a) Schematic representation of the AP-DBD reactor setup and (b) traces of the voltage pulse and current discharges. B

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules were coated with a 5 nm platinum film prior to the SEM observations. The thicknesses were confirmed by a KLA-Tencor P-17 Stylus profiler (Milpitas, CA). The mass of the films was determined by weighing the substrate before and after deposition using a Sartorius ME36S scale (Goettingen, Germany). The growth rates in thickness and weight were determined from the measured thicknesses and masses and the time spent in the deposition area, which is related on the high-voltage electrodes width (30 mm), the treatment speed (1 mm s−1), and the number of passes. Chemical Analysis. FTIR transmission measurements were performed on a Bruker Vertex 70 spectrometer (Ettlingen, Germany) and measured with a MCT detector. All FTIR data were normalized according to the CO stretching band measured at 1730 cm−1 as it should be less altered by this level of plasma discharge energy26,36 and is unaffected by cross-linking reactions. XPS measurements were performed on a Kratos Axis-Ultra DLD instrument, using an Al Kα source (1486.6 eV) with a pass energy of 20 eV and an energy resolution of 0.5 eV. A flooding gun was used to reduce charging effect on the samples surface. The polymers chain growth was determined by size exclusion chromatography using a Thermo Scientific (Sunnyvale, CA) Dionex UltiMate 3000 LC system. Each measurement corresponds to the soluble fraction of the polymer in THF and not

necessarily the entire distribution. The elution times were translated into molecular weights using polystyrene standards ranging from 162 to 364 000 g mol−1 (Agilent EasiVial PS-M). MALDI-HRMS measurements were performed on an AP-MALDI PDF+ ion source from MassTech Inc. (Columbia, MA) coupled to a LTQ/Orbitrap Elite from Thermo Scientific (San Jose, CA). Both dihydroxybenzoic acid and dithranol were tested as the matrix due to their known efficiency for the ionization of PMMA37 and polyacrylate plasma polymers.31 The best results were obtained from dithranol diluted in THF, possibly due to its hydrophobicity. 0.1 μL of matrix was then spotted on the thin films to form a cocrystal with the soluble part of the film when the THF evaporates.



RESULTS AND DISCUSSION To gain insight into the mechanisms influencing the formation of polymer thin films when using ultrashort square pulse dielectric barrier discharge, three series of thin films have been deposited from a set of three different methacrylate monomers, i.e., MMA, BMA, and GMA, with different kinetic rate constants, i.e., kp and kt (Table 1). To ensure the accuracy of the comparisons made along the present work, the carrier flows have been adapted according to the saturated vapor pressure (Psat) of each of the investigated monomers in order to deliver the same rate of molecules to the deposition zone (Table S1). For each of the prepared series, a constant 15 μs square-wave pulse electrical excitation, generating two distinct current discharges (ca. 50 ns) at the voltage rising and falling edges (Figure 1b), was employed. The repetition frequency of the high voltage ultrashort square-wave pulse was tuned between 10 and 10 000 Hz, allowing to investigate plasma off-times (toff) from 100 ms to 100 μs, respectively. Interestingly, naked eye observation readily allowed to identify color and physical state differences between each of the prepared thin films (Figure S2). Under strictly identical deposition conditions (i.e., same HV pulse’s frequency and same monomer flow rate), the physical state of the thin films was strongly influenced by the nature of the methacrylate monomer. While homogeneous and solid organic layers were grown for frequencies in the range from 100 to 1000 Hz from MMA and GMA (Figure S2b,c), viscous

Table 1. Kinetic Properties and Skeletal Formula of Each Methacrylate Monomer Describing Their Adsorption Ability (Psat) and Their Polymeric Growth (kp and kt) as Well as Their Chemical Functionalities

*

kt value for GMA could not be obtained from the literature.

Figure 2. Weight and thickness increment per discharge according to the cycle time (ton + toff) and growth rates per second according to the duty cycle for the films grown from (a, b) MMA, (c, d) BMA, and (e, f) GMA. C

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules and nonhomogeneous films were obtained from BMA (Figure S2a). Beyond 1000 Hz, rough and solid plasma-polymer coatings were grown for all monomers. Such discrepancies are indicative of variation in the chemistry of the films. In addition, obvious growth rate differences, assumed from the color variations38 (Figure S2b,c), were observed and suggest both an influence of the methacrylate monomers (e.g., kinetic properties, functional groups) and the HV pulse’s frequency. Effect on the Growth Rates. Following to the initial observations described above, the weight and thickness growth rates of the films were evaluated from mass and profilometry measurements. Such as reported in a previous work,31 higher thickness and weight increments per cycle (i.e., ton + toff, with ton kept constant) were observed for longer toff (Figure 2a,c,e), highlighting the occurrence of deposition reactions during toff. In spite of exhibiting similar trends, the weight and thickness increment per cycle plots are not identical for all the methacrylate monomers. Indeed, the plateau indicating the termination of the free-radical polymerization mechanism is reached for toff ranging from 10 to 100 ms for the MMA and GMA,31 respectively. On the other hand, the plots of the thickness and weight growth rates (Figure 2b,d,f) according to the duty cycle (DC = ton/(ton + toff)) highlight their strong dependency on the plasma pulses frequency and indicate the occurrence of different mechanisms (i.e., free radical polymerization, plasmapolymerization or random cross-linking, dissociation, ionization, etching) during the PECVD of methacrylate monomers. In accordance with our previous observation, for values of toff above 1 ms (i.e., DC below 0.01%) the deposition rates quickly decay due to the extinction of the plasma activated species and the progressive termination of the free-radical polymerization process. In other words, for low duty cycles (i.e., long toff), the concentrations of initiating species and free radicals have too much decreased, explaining the drop in growth rate for low DC. For high DC (i.e., short toff), the plasma processes are anticipated to overcome the free-radical polymerization processes since each molecule will be exposed to a higher number of plasma discharges before reaching the substrate. Thus, an increase of chains termination is expected due to the high production of radicals and other species that could interrupt the growth of the polymeric chains generating short chain oligomers. As the DC is increased, the fragmentation and dissociation of the monomer is more pronounced, leading to a drop of the deposition rates for the BMA and GMA cases (Figure 2d,f). The DC for which this drop occurs is dependent on the monomer, with maximum growth rates observed at lower DC for GMA (i.e., 0.01%) than for BMA (i.e., 0.03%) and MMA (≥0.1%). In addition, while the densities of the MMA and BMA thin films, ca. 1.21 and 1.59 g mL−1, respectively, are pretty constant over the range of studied frequencies, the growth rates in weight and thickness of the GMA thin films do not follow the same trend. Such change of density (Figure 2f) may arise from the presence of the reactive epoxy ring of GMA, which is expected to be more sensitive toward plasma fragmentation for high DCs and would favor a cross-linking by creation of a radical anchor point through ringopening. From the above growth rates observations, it is clear that the films elaborated from the AP-PECVD of methacrylate monomers are the result of a competition between the plasma polymerization and the conventional free-radical polymerization mechanisms. This competition appears to be dependent on the monomer such as observed from the maximum deposition rates per cycle (Figure 2a,c,e) and the shift of the frequency for which the maximum deposition rate is obtained (Figure 2b,d,f).

Figure 3. FTIR spectra of the thin films grown from MMA, BMA, and GMA at different discharge frequencies. The spectra of both the liquid precursor (MMA, BMA, and GMA) and the conventionally polymerized PMMA, PBMA, and PGMA powder are shown as references on the corresponding figure. The CO band39 (1723 cm−1), present in the methacrylate monomer, reference polymer, and the prepared thin films, is used to normalized all the spectra.

In any case, the free-radical polymerization pathway is always greatly favored for the lowest DCs (i.e., the longest toff). During these long plasma off-times, the occurring surface reactions are assumed to be close to the radical polymerization steps taking place in both the bulk phase and in an iCVD process. Thus, for very long toff, the deposition rate should be strongly influenced by the concentration of initiating species and active growing chains ([M•]), the concentration of monomer adsorbed on the surface ([M]), and the rate constant for propagation (kp). The monomer surface concentration being directly related to the ratio of the monomer’s vapor-phase concentration (PM) and its D

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

The two following sections aim at evaluating the monomer’s chemistry retention and conventional polymerization, which are assuredly the main weaknesses of plasma polymers. FTIR was employed to investigate the different bonding arrangements in the films and compare them to their precursor and corresponding reference polymer powder (Figure 3). Regardless of the pulse’s frequency or thin films precursor, the vinyl bonding vibrations39 (1635 and 1310 cm−1) disappeared for all the grown thin films, supporting its reduction along the PECVD process. Irrespective of the monomer investigated, the spectra of the films obtained for the lowest frequencies display a very close match in comparison to reference polymers, while a greater number of discrepancies are observed as the pulse’s frequency gets higher. Typically, a rise is observed in the intensity of the vibrational bands corresponding to CH2 bonding39 (2953, 2875, and 1447 cm−1), opposed to a drop for of the CH3 vibrational bands domains39 (2995, 2840, 1483, and 1391 cm−1). An increase in the carbon bonds substitutions is indicative of a more pronounced reorganization of the monomers molecules toward a cross-linked structure. An interesting disparity appears among the polymers’ vibrational energies between 3200 and 3600 cm−1, corresponding to hydrogen stretching vibrations. While the ppMMA and ppBMA FTIR spectra show a quick surge for hydroxyl bands vibrations39,40 (3252 and 3427 cm−1) as the pulse’s frequency decreases, ppGMA exhibit the opposite behavior with an increase of the hydroxyl stretching vibrations for high pulse’s frequencies. If the opening of the sensitive epoxy function of ppGMA at high DCs seems to be the obvious reason for the increase of OH vibration, the antagonist observation seemed more subtle to understand for the two other methacrylate monomers. A possible explanation to this effect could be the strong influence of intermolecular vibrations39,40 due to the addition of proton acceptor groups,41 leading to an important broadening of the hydroxyl band toward lower energy levels (3252 cm−1).42 The presence of proton acceptors originating from the surrounding atmosphere (despite the surrounding atmosphere passivation using Ar barrier fluxes) is a fair

saturated vapor pressure (Psat), the studied methacrylate monomer surface concentrations can be ranked as [MMMA](ad) < [MBMA](ad) < [MGMA](ad) as PM is kept constant for all the deposition experiments. Additionally, their respective propagation rate constants are ordered as follows: kpMMA ∼ kpBMA < kpGMA. Following those observations, it is interesting to note that the deposition rates (DR) for the low-frequency experiments (100 Hz in the following comparison), which ranked as followsDRMMA (0.18 nm s−1) < DRBMA (0.24 nm s−1) ≪ DRGMA (2.50 nm s−1)are in accordance with the [M] and kp ratings, supporting the assumption that the surface reactions taking place during the plasma off-times are analogous to the radical polymerization steps in the bulk phase. Effect on the Chemistry. Such as expressed in the previous section, different mechanisms (e.g., free radical polymerization, plasma polymerization or random cross-linking, dissociation, ionization, etching) occur during the PECVD of methacrylate monomers. In addition to influencing the growth rates, the combination of these mechanisms is well-known to impact the chemical composition and structure of the resulting films. Table 2. Atomic Ratios of ppMMA, ppBMA, and ppGMA Thin Films at Different Discharge Frequencies in Comparison to Their Theoretical Values ppMMA

ppBMA

ppGMA

theory 100 Hz 1000 Hz 10000 Hz theory 100 Hz 1000 Hz 10000 Hz theory 100 Hz 1000 Hz 10000 Hz

O (%)

N (%)

C (%)

28.5 27.3 25.1 19.0 20.0 20.9 17.6 16.6 30.0 26.6 26.7 13.8

0.0 0.1 0.1 0.2 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.8

71.5 72.6 74.9 80.8 80.0 79.1 82.5 83.3 70.0 73.4 73.3 85.4

Figure 4. XPS spectrum of the O 1s and C 1s core levels for the ppMMA, ppBMA, and ppGMA films deposited at different pulse’s frequencies, with the C−H binding energy fixed at 284.8 eV. The O 1s and C 1s core level of the conventionally polymerized PMMA, PBMA, and PGMA powders are shown as references. E

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules assumption; however, it assuredly requires more investigations to draw any conclusion regarding its relationship with the frequency of pulse. XPS elemental analysis of the films elaborated using low pulse’s frequencies display almost no difference with the theoretical values for their corresponding polymer (Table 2). In addition, no evidence of nitrogen contamination from the surrounding atmosphere was detected for these films. On the other hand, higher carbon concentrations and lower oxygen concentrations than the expected ones were measured for the films elaborated from higher DCs. Furthermore, trace levels of nitrogen, integrated from excessive reaction of the monomers with the surrounding atmosphere, were detected. The C 1s and O 1s core levels envelopes for the thin films described in the present work were compared in order to highlight both the similarities and changes in chemistry (Figure 4). Similarly to the FTIR observations, the C 1s and O 1s core levels spectra for the films obtained for the lowest frequencies display a close match to the ones for the reference polymers. While increasing the DC, C−O simple bonds43 (i.e., C3 and O2) appear as more prone to alteration. In particular, the GMA’s C−O components,43 including both the ether and the epoxy bonds (i.e., C3, O2 and C5, O3, respectively) were shown to almost completely disappear at 10 000 Hz. Effect on the Polymeric Growth. Eventhough plasmabased processes are very unlikely to ever produce true homopolymers due to the highly reactive and nonspecific nature of plasmas, from both the previous FTIR and XPS observations, the chemical functionalities appear to be very well preserved for low discharge frequencies (i.e., below 100 Hz). This contrasts with the usual cross-linking and reactive functions breakdown observed in plasma polymers. The following section will aim at demonstrating the formation of conventional polymer layers from specific polymer characterization methods, i.e., matrixassisted laser desorption/ionization high resolution mass spectrometry (MALDI-HRMS) and size exclusion chromatography (SEC). Figure 5 shows the mass spectra for the films elaborated from GMA at 100, 1000, and 10 000 Hz. The mass spectra in the range m/z = 100−1400 of the films elaborated from the lowest frequency are dominated by PGMA proton adduct with proton terminal groups [H(GMA)nH + H]+ up to n = 7 repeating units (Figure 5a). The first modification, with a mass difference of 0.0364 Da, corresponds to the substitution of a CH4 unit (i.e., a terminal methyl bond) with an O unit (i.e., a ketone/aldehyde bond) hinting on an effect of the atmospheric environment (O2 and H2O). This reaction appears to be even more prominent for the films deposited from MMA (Figure 6b) and BMA (Figure 7b) at low DC, correlating with an increase of hydroxyl function and the previous FTIR observations for the 3252 and 3427 cm−1 bands (Figure 3). As the pulse’s frequency is increased, a greater number of structural modifications are observed, indicating a prevailing influence of the plasma polymerization processes, including their side reactions (Figures 5a, 6a, and 7a). Thanks to the high mass resolution, zooming into one of the main oligomer unit (Figures 5b, 6b, and 7b) reveals the variations originating from the plasma phase as numerous peaks caused by fragmentation are found for the films elaborated from the highest frequency, resulting from one to several dissociation and recombination reactions. However, the number of species between the oligomer units appears to be extremely monomer dependent, with MMA films (Figure 6a) displaying a non-negligible amount in comparison to BMA (Figure 6a) and especially GMA (Figure 5a) films.

Figure 5. MALDI-HRMS spectra in the mass ranges (a) m/z = 100−1400 and (b) m/z = 428.9−429.6 for the films elaborated from GMA for various pulse’s frequencies.

Those observations would have an important consequence on the growth mechanism as it would suggest that MMA reactive species are more sensitive to side-reactions while BMA and GMA growing chains tend to quickly adsorb on the substrate and deactivate, relating to their vapor pressure Psat, before being reactivated by either the plasma or an incoming radical monomer, hence statistically lowering the addition of impurities. In order to understand the effect of these plasma formed species on the polymeric chain growth, SEC measurements were performed on all the grown films. The measurements were plotted and compared to each other in elution time since the molar weight cannot be reliably assessed due to the chemistry discrepancies between the low and high pulse’s frequency thin films. The organic layers elaborated from low DCs display one main peak (i.e., 9 min for ppMMA and ppBMA and 6.5 min for ppGMA) (Figure 8a,c,e), which can be seen as thealmost uniquecontribution of a singular distribution, implying the existence of one predominant compound and suggesting a polymer-like growth. In addition, F

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 6. MALDI-HRMS spectra in the mass ranges (a) m/z = 280−1400 and (b) m/z = 401.0−401.42 for the films elaborated from MMA for various pulse’s frequencies.

Figure 7. MALDI-HRMS spectra in the mass ranges (a) m/z = 100−1400 and (b) m/z = 427.0−427.6 for the films elaborated from BMA for various pulse’s frequencies.

a low-intensity peak or plateau appears at shorter retention times (i.e., higher molecular weights) for both MMA and BMA monomers (Figure 8b,d), indicating the potential formation of polymer compounds up to several hundreds of thousands of grams per mole. Assuming that polymeric chains of PMMA, PBMA, and PGMA are grown for the low DCs, average molar masses in weight (Mw) of 2596, 4761, and 94 364 g mol−1, respectively, were estimated for the methacrylate films grown at 100 Hz, integrated around their polymer distribution peaks (i.e., between 5 and 10.5 min for PMMA and PBMA and narrowed around 5.6 and 8 min for PGMA due to the low intensity/ baseline ratio). On the other hand, several overlapping peaks are observed for the high pulse’s frequency plasma-polymer coatings, suggesting the presence of several compounds with different chemistries. Surprisingly, the elution times for these peaks are similar to the ones observed for the low DCs polymer layers grown from MMA and BMA. Thus, the arising distribution peaks (ca. 9 min) could easily be misinterpreted as long polymeric chains comparable to the ones observed for low DCs. However, based on our previous FTIR and XPS investigations,

high DCs are proven to lead to the formation of altered chemistries with a high degree of cross-linking. These highly cross-linked structures are expected to rapidly migrate through the chromatographic column due to their 3D conformation and give rise to the short elution time peaks observed. Hence, the formed chemistries cannot be calibrated from any known compounds, as they would incorrectly appear as higher molecular weight species. In the case of GMA, several overlapping peaks are also observed for the plasma-polymer coating grown at 10 000 Hz at an elution time comparable to the ones observed for the films elaborated from MMA and BMA at 10 000 Hz (ca. 9 min), suggesting that for higher DCs, the deposition is more driven by the plasma condition rather than by the monomer’s intrinsic kinetics properties. Interestingly, for the plasma-polymer coating grown from GMA at 10 000 Hz, a distribution at an identical elution time as the polymer distribution of the 100 Hz polymer layers is observed (ca. 6.5 min), indicating the remanence of the free-radical polymerization path. Further, taking in consideration that the plasma residues elution time coincides with PMMA and PBMA polymeric distribution, we can relate the alterations of the low tail of the Gaussian distribution G

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 8. SEC measurement for the films elaborated from MMA, BMA, and GMA for various pulse’s frequencies: (a, c, e) whole range spectra and (b, d) focus on the high molecular weight tails when observable.

(Figure 8b) to the evolution of polymer favoring mechanisms (high molecular plateau) as well as the plasma polymerization favoring mechanisms (expansion of the fragment distribution peaks) over the frequency. Therefore, the competition between the “kinetic regime”, prominent for the lowest frequencies, and the “plasma polymer regime”, prominent for the highest frequencies, is readily observed from the SEC spectra of the GMA samples with a greater fraction related to the Gaussian polymeric distribution at low frequency and the apparition of new distributions related to plasma-polymer-like compounds at higher frequencies (Figure 8e). Such a transition could not be readily evidenced from the SEC spectra of the MMA and BMA samples due to the overlap of the polymeric and plasmapolymer distributions (Figure 8a,c). Thus, the apparent average molar mass in weight (Mw*) was determined at all frequencies for MMA and BMA distributions, taken between 5 and 10.5 min to remove the influence of the monomer and solvent contamination. For each of the investigated monomer, the apparent Mw* plotted versus the DC effectively exhibits the transition from one dominant mechanism to the other, displaying a bowl shape with a minimum around 1000 Hz and an increase of the apparent Mw* for both low pulse’s frequencies (i.e., long toff) due to the promotion of a “kinetic regime” and at high pulse’s frequencies (i.e., short toff) as it converts toward a “plasma polymer regime” (Figure 9). Effect on the Morphology. Deposition rates (Figure 2) and the apparent average molar mass in weight (Figure 9) both highlighted the occurrence of a competition between the plasma-polymerization and the conventional free-radical polymerization mechanisms during the pulsed AP-PiCVD of methacrylates. The influence of the pulse’s frequency and the methacrylate monomer on this competition can also be observed on the morphology of the films (Figures 10 and 11). While the top view SEM observations reveal the formation of smooth and particle-free films for the lowest pulse’s frequencies, rough surfaces were formed for frequencies above 10 000 Hz for MMA and BMA and 1000 Hz for GMA (Figure 10).

Figure 9. Apparent average molar mass in weight (Mw*) for the films elaborated from MMA and BMA for various pulse frequencies. Mw* values are extracted from the respective SEC measurements and a polystyrene calibration. One should note that the Mw* are given for the need of the discussion and that these values do not constitute an accurate information for the high DC films. Because of the splitting of the polymeric and plasma-polymer distributions for the GMA samples, the integration of both Gaussians in a single Mw* is not required and not possible.

Interestingly, the morphology was shown to be independent of the growth rate, implying that the film’s roughening is not related to an excessive deposition rate but rather to excessive gas phase reactions induced by the highest pulse’s frequency. A concrete example is the SEM images of the films elaborated from GMA at 100 Hz and 10 000 Hz, which exhibit strong discrepancies in their morphology in spite of identical growth rates. In a perfectly repeatable manner, the thin films realized at pulse’s frequencies strictly inferior to the deposition rate transition display a smooth surface (Figure 10). At the transition frequency, several rough features start to emerge on the surface, and above this transition, a greater number of particles is H

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 10. (left) Growth rates per second according to the duty cycle for the films grown from MMA, BMA, and GMA. (right) Top-view SEM images of the ppMMA, ppBMA, and ppGMA films at 100, 1000, and 10 000 Hz.

thin films with a high level of conformality when using low PM/Psat values (Figure 11). Such as for iCVD, the use of a vinyl monomer is a prerequisite and can allow the simultaneous synthesis and deposition of polymer with various functionalities.32,45 While iCVD implies the use of an initiator (e.g., di-tert-butyl peroxide) broken down into radicals by hot filaments to begin the free-radical polymerization of vinyl monomers at the substrate surface, our approach involves the ultrashort plasma pulses to initiate the free-radical polymerization reaction. By replicating the free-radical polymerization in the vapor phase at atmospheric pressure, the described method provides a simple one-step and up-scalable route toward the deposition of a wide range of polymer thin films.46 Interestingly, the method can easily be adapted to existing AP-DBD surface treatment lines through the implementation of a nanosecond HV pulses generator. The present study highlighted several times the duality between two different modes (Figure 12), i.e., the plasma-enhanced CVD (PECVD) and the plasma-initiated CVD (PiCVD) of plasma-polymer and polymer-like thin films, respectively. Each of the two modes, both occurring while combining plasma and vinyl monomers, can be greatly favored under certain conditions. A quite obvious rule when performing the plasma-initiated chemical vapor deposition (PiCVD) of polymer thin films is the use of a low repetition frequency of the ultrashort square-wave pulses. This repetition frequency has to be consistent with the lifetime of the free-radical polymerization for the selected monomer. Indeed, the weight and thickness increment per discharge according to the cycle time (ton + toff) revealed that in the case of MMA monomer no significant deposition occurs when toff exceeds 10 ms (Figure 2a). Thus, there is no point using frequencies lower than 100 Hz under these conditions as it would only result in a decay of the growth rate with no further enhancement of the chemistry. In contrast, the weight and thickness increment per discharge according to the cycle time demonstrated that the free-radical polymerization of the GMA

formed as the pulse’s frequency increases, rapidly deteriorating the film morphology down to a highly granular surface. Such behavior is related to the gas phase reaction of the methacrylate monomers with the abundant reactive species generated by the plasma discharges that induce a high concentration of condensable vapors, increasing the nucleation and growth of numerous particles. To confirm the above statements, a polymer layer elaborated at 100 Hz and a plasma-polymer coating at 10 000 Hz were deposited on trenched silicon wafers. The SEM cross-section observation of the thin film deposited from GMA at 10 000 Hz reveals the nonconformality of the coating, with the absence of deposition after a few hundred nanometers down into the trench. A bridge between the two sides of the trench is even created. Such comportment is not surprising as PECVD processes are well-known for their nonconformal behavior due to the high sticking coefficient of the formed reactants that readily condensate on the mouth of porosities. On the other hand, the smooth polymer layers elaborated from MMA, BMA, or GMA26 at 100 Hz exhibit an excellent conformality with a ratio between the thicknesses at the bottom and top of the trench higher than 85%. For the low pulse’s frequencies, the methacrylate monomers adsorb to the surface where they are consumed by the free-radical polymerization reaction with the previously adsorbed radicals generated by the ultrashort square-wave pulse plasma discharges. Guidelines to PiCVD Polymerization. Although plasmabased processes, including AP-PiCVD, are unlikely to ever produce true homopolymers thin films such as iCVD does, the present investigations assuredly exhibit strong similarities with the iCVD method13,44 with respect to the growth mechanisms and the resulting thin films’ characteristics. For low plasma pulse’s frequencies, surface reaction mechanisms (i.e., adsorption and subsequent free-radical polymerization) are preponderant, leading to an unprecedented degree of polymerization for PECVD processes (Figure 8) coupled to the formation of I

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 12. Apparent average molar mass in weight (Mw*) and growth rates per second according to the duty cycle for the films grown from (top) MMA and (bottom) BMA.

carrier gas flux), bearing in mind that plasma processes are strongly affected by the composition of the plasma gas.49 An alternative option is to decrease the substrate temperature in order to enhance the adsorption rate and favor the polymeric growth. It is worth noting here that in contrast with several other PECVD processes,50 the use of low-frequency ultrashort squarewave pulse plasma DBD does not lead to an increase of the substrate temperature. Moreover, early trials of MMA and BMA films grown with low-frequency modes all yielded highly viscous and nonregular coatings until inert gas curtains (e.g., argon) were added on each side of the deposition area (Figure S1), highlighting the importance of passivating the surrounding open-air atmosphere to avoid the non-negligible contamination and chain termination in accordance with our MS observations (Figures 5b, 6b, and 7b). In addition to these few guidelines, the monomer chemical structure and intrinsic properties are expected to influence the PiCVD of polymer-like thin films. Notably, the propagation and termination rate constants (kp and kt) have been shown to influence the deposition kinetics. Furthermore, MALDI-HRMS measurements revealed several alterations that are dependent on the selected monomers, and deeper investigations of the mass spectrometry data are currently under progress to gain a better understanding of the behavior of each of the studied monomers.

Figure 11. Cross-section SEM images of the thin films deposited from MMA at 100 Hz and GMA at 10 000 Hz on 5 μm deep and 600 nm wide trenches on silicon wafers. It is noteworthy to mention that the conformal or nonconformal behavior of the thin films is mainly related to the pulse’s frequency and to a lesser extend to the PM/Psat ratio. Polymer layers grown from GMA at low pulse’s frequency also displayed an excellent conformality.26

monomer can perpetuate over 100 ms31 (Figure 2e), suggesting that a 10 Hz discharge frequency would lead to an improvement of the film’s chemistry. Such as described previously, for low DC (i.e., low frequencies), the deposition rate is shown directly related to the PM/Psat ratio (Figure S3) and hence to the volume of multilayers adsorbed through a BET isotherm.47 Thus, for a monomer with a lower vapor pressure (Psat), adsorption to the surface would be promoted, and both the growth rate and the conventional polymerization pathway will be enhanced. Nevertheless, the vapor pressure should be sufficient enough to allow delivery of the monomer at atmospheric pressure. Aerosol-assisted CVD is a common practice at atmospheric pressure when employing low-Psat precursors;48 however, this is not recommended for the PiCVD of polymer-like thin films as this would result in many condensation or droplet spots and form irregular and nonconformal coatings. For monomer with high vapor pressure, one could consider increasing PM (i.e., through the increase of the



CONCLUSION A set of three methacrylates monomers (i.e., MMA, BMA, and GMA) have been investigated to gain a deeper understanding of the mechanisms driving our novel atmospheric-pressure plasma-initiated chemical vapor deposition (AP-PiCVD) process. Thanks to extremely short plasma discharges (i.e., 100 ns) and J

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Macromolecules long plasma off-time (i.e., 100 μs to 31.6 ms) (Figure 1b), the fragmentation and recombination mechanisms encountered in PECVD were strongly attenuated, while the self-sustained freeradical polymerization pathway was strongly promoted. In a common fashion, low frequencies were shown to favor surface reactions, such as low-pressure CVD approaches toward the formation of polymer layers (e.g., iCVD), yielding highly conformal thin films. FTIR and XPS investigations highlighted the excellent chemical retention of the formed layers. Quite impressively, SEC measurements displayed average molecular masses in weight up to 94 000 g mol−1 for the layers grown from GMA. Such achievement is an unprecedented value for both plasma and atmospheric CVD processes. While SEC investigations allowed to monitor changes in molecular distribution, indicating a progressive transition from a “conventional” to a “plasma-polymer” chemistry shared by all the investigated methacrylate monomers, their average molecular weights were extremely contrasted, especially when comparing GMA to both BMA and MMA. This difference appears related to the monomer’s sensitivity to plasma fragmentation and surrounding atmospheric species integration. On the other hand, as the HV pulse frequency was increased, each monomer appeared to have its own sensitivity toward the plasma-induced reactive species available in the gas phase, displaying individual thresholds transitioning the dominant mechanisms between surface and gas phase reactions such as evidenced by top-view SEM imaging, hence relating to their adsorption capacities. AP-PiCVD have consequently shown to be a promising process that overcame the main drawbacks from usual PECVD. As such, these findings may be beneficial to the wide range of practical applications of thin polymer films, ranging from the large scale applications on paper and textile substrates to smaller scale applications on micro- and nanostructured devices.44 In particular, the PiCVD of thermoresponsive polymers is currently explored.





ABBREVIATIONS



REFERENCES

CVD, chemical vapor deposition; AP, atmospheric pressure; PECVD, plasma-enhanced CVD; PiCVD, plasma-initiated CVD; MMA, methyl methacrylate; BMA, butyl methacrylate; GMA, glycidyl methacrylate; FTIR, Fourier transform infrared; XPS, X-ray photoelectron spectroscopy; SEC, size exclusion chromatography; SEM, scanning electron microscopy; MALDIHRMS, matrix-assisted laser desorption/ionization high resolution mass spectrometry.

(1) Liu, J.; Li, R. C.; Sand, G. J.; Bulmus, V.; Davis, T. P.; Maynard, H. D. Keto-Functionalized Polymer Scaffolds as Versatile Precursors to Polymer Side-Chain Conjugates. Macromolecules 2013, 46, 8−14. (2) Stejskal, J.; Bober, P.; Trchová, M.; Kovalcik, A.; Hodan, J.; Hromádková, J.; Prokeš, J. Polyaniline Cryogels Supported with Poly(vinyl alcohol): Soft and Conducting. Macromolecules 2017, 50, 972. (3) Chen, N.; Kim, D. H.; Kovacik, P.; Sojoudi, H.; Wang, M.; Gleason, K. K. Polymer Thin Films and Surface Modification by Chemical Vapor Deposition: Recent Progress. Annu. Rev. Chem. Biomol. Eng. 2016, 7, 373−393. (4) Stuart, M. a C.; Huck, W. T. S.; Genzer, J.; Müller, M.; Ober, C.; Stamm, M.; Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V.; Urban, M.; Winnik, F.; Zauscher, S.; Luzinov, I.; Minko, S. Emerging applications of stimuli-responsivepolymer materials. Nat. Mater. 2010, 9, 101−113. (5) Singha, K. A Review on Coating & Lamination in Textiles: Processes and Applications. Am. J. Polym. Sci. 2012, 2, 39−49. (6) Hassan, M. M.; McLaughlin, J. R. Formation of Poly(methyl methacrylate) Thin Films onto Wool Fiber Surfaces by Vapor Deposition Polymerization. ACS Appl. Mater. Interfaces 2013, 5, 1548−1555. (7) Roland, S.; Gamys, C. G.; Grosrenaud, J.; Boissé, S.; Pellerin, C.; Prud’Homme, R. E.; Bazuin, C. G. Solvent Influence on Thickness, Composition, and Morphology Variation with Dip-Coating Rate in Supramolecular PS-b-P4VP Thin Films. Macromolecules 2015, 48, 4823−4834. (8) Ogawa, H.; Takenaka, M.; Miyazaki, T.; Fujiwara, A.; Lee, B.; Shimokita, K.; Nishibori, E.; Takata, M. Direct Observation on SpinCoating Process of PS-b-P2VP Thin Films. Macromolecules 2016, 49, 3471−3477. (9) Gleason, K. K. CVD Polymers Fabrication of Organic Surfaces and Devices; Gleason, K. K., Ed.; Wiley-VCH: 2015; p 1. (10) Tenhaeff, W. E.; Gleason, K. K. Initiated and Oxidative Chemical Vapor Deposition of Polymeric Thin Films: iCVD and oCVD. Adv. Funct. Mater. 2008, 18, 979−992. (11) Reeja-Jayan, B.; Kovacik, P.; Yang, R.; Sojoudi, H.; Ugur, A.; Kim, D. H.; Petruczok, C. D.; Wang, X.; Liu, A.; Gleason, K. K. A Route Towards Sustainability Through Engineered Polymeric Interfaces. Adv. Mater. Interfaces 2014, 1, 1400117. (12) Asatekin, A.; Barr, M. C.; Baxamusa, S. H.; Lau, K. K. S.; Tenhaeff, W.; Xu, J.; Gleason, K. K. Designing polymer surfaces via vapor deposition. Mater. Today 2010, 13, 26−33. (13) Lau, K. K. S.; Gleason, K. K. Initiated Chemical Vapor Deposition (iCVD) of Poly(alkyl acrylates): An Experimental Study. Macromolecules 2006, 39, 3688−3694. (14) Scheltjens, G.; Da Ponte, G.; Paulussen, S.; De Graeve, I.; Terryn, H.; Reniers, F.; Van Assche, G.; Van Mele, B. Deposition Kinetics and Thermal Properties of Atmospheric Plasma Deposited Methacrylate-Like Films. Plasma Processes Polym. 2016, 13, 521−533. (15) Kakaroglou, A.; Nisol, B.; Baert, K.; De Graeve, I.; Reniers, F.; Van Assche, G.; Terryn, H. Evaluation of the Yasuda parameter for the atmospheric plasma deposition of allyl methacrylate. RSC Adv. 2015, 5, 27449−27457. (16) Tendero, C.; Tixier, C.; Tristant, P.; Desmaison, J.; Leprince, P. Atmospheric pressure plasmas: A review. Spectrochim. Acta, Part B 2006, 61, 2−30.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.7b00461. Table summarizing the deposition conditions; optical images of the thin films grown at 100 Hz; voltage and current traces of the ultrashort square pulses; experimental BET isotherm plot and its theoretical fit (PDF)



Article

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]; Tel +352 275 888 578 (N.D.B.). ORCID

Nicolas D. Boscher: 0000-0003-3693-6866 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The Luxembourg National Research Fund (fnr.lu) is thanked for financial support through the NANOPOLYPULSE project (C14/MS/8345246). C. Vergne and Dr. J. Bour from LIST are acknowledged for insightful discussions and acquisition of the SEC and MS measurements. We thank O. Bouton, D. Abessolo Ondo, A. Combrisson, and Dr. J. Guillot from LIST for the assistance with the AP-DBD reactor and XPS analysis. K

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Atmospheric Pressure Dielectric Barrier Discharge. Plasma Processes Polym. 2010, 7, 163−171. (34) Tarducci, C.; Schofield, W. C. E.; Badyal, J. P. S.; Brewer, S. A.; Willis, C. Monomolecular Functionalization of Pulsed Plasma Deposited Poly(2-hydroxyethyl methacrylate) Surfaces. Chem. Mater. 2002, 14, 2541−2545. (35) Wetzel, S. J.; Guttman, C. M.; Girard, J. E. The influence of matrix and laser energy on the molecular mass distribution of synthetic polymers obtained by MALDI-TOF-MS. Int. J. Mass Spectrom. 2004, 238, 215−225. (36) Mavroudakis, E.; Cuccato, D.; Moscatelli, D. On the Use of Quantum Chemistry for the Determination of Propagation, Copolymerization, and Secondary Reaction Kinetics in Free Radical Polymerization. Polymers 2015, 7, 1789−1819. (37) Ogo, Y.; Kyotani, T. Effect of Pressure on the Termination Rate Constant in Free Radical Polymerization. Correlation between Rate Constant and Monomer Viscosity. Makromol. Chem. 1978, 179, 2407−2417. (38) Macleod, H. G. Thin-Film Optical Filters; Institute of Physics Publishing: 2001; p 500. (39) Socrates, G. Infrared and Raman characteristic group frequencies; Wiley-VCH: 2004. (40) Bellamy, L. J. The Infrared Spectra of Complex Molecules; Chapman and Hall: 1980; p 254. (41) Baitinger, W. F.; Schleyer, P. vo. R.; Murty, T. S. S. R.; Robinson, L. Nitro Groups as Proton Acceptors in Hydrogen Bonding. Tetrahedron 1964, 20, 1635−1647. (42) Gordy, W. Spectroscopic Evidence of Hydrogen Bonds: Aniline and Some Substituted Phenols. J. Chem. Phys. 1939, 7, 167−171. (43) Beamson, G.; Briggs, D. High Resolution XPS of Organic Polymers; Wiley-VCH: 1992. (44) Wang, M.; Wang, X.; Moni, P.; Liu, A.; Kim, D. H.; Jo, W. J.; Sojoudi, H.; Gleason, K. K. CVD Polymers for Devices and Device Fabrication. Adv. Mater. 2017, DOI: 10.1002/adma.201604606. (45) Bonot, S.; Mauchauffé, R.; Boscher, N. D.; Moreno-Couranjou, M.; Cauchie, H. M.; Choquet, P. Self-Defensive Coating for Antibiotics DegradationAtmospheric Pressure Chemical Vapor Deposition of Functional and Conformal Coatings for the Immobilization of Enzymes. Adv. Mater. Interfaces 2015, 2, 1500253. (46) Mauchauffé, R.; Bonot, S.; Moreno-Couranjou, M.; Detrembleur, C.; Boscher, N. D.; Van De Weerdt, C.; Duwez, A. S.; Choquet, P. Fast Atmospheric Plasma Deposition of Bio-Inspired Catechol/Quinone-Rich Nanolayers to Immobilize NDM-1 Enzymes for Water Treatment. Adv. Mater. Interfaces 2016, 3, 1500520. (47) Atkins, P. Physical Chemistry; Oxford University Press: 2006; p 920. (48) Knapp, C. E.; Parkin, I. P.; Carmalt, C. J. Aerosol-Assisted Chemical Vapor Deposition of Transparent Conductive GalliumIndium-Oxide Films. Chem. Mater. 2011, 23, 1719−1726. (49) Kobayashi, H.; Bell, A. T.; Shen, M. Plasma Polymerization of Saturated and Unsaturated Hydrocarbons. Macromolecules 1974, 7, 277−283. (50) Boscher, N. D.; Olivier, S.; Maurau, R.; Bulou, S.; Sindzingre, T.; Belmonte, T.; Choquet, P. Photocatalytic anatase titanium dioxide thin films deposition by an atmospheric pressure blown arc discharge. Appl. Surf. Sci. 2014, 311, 721−728.

(17) Massines, F.; Sarra-Bournet, C.; Fanelli, F.; Naudé, N.; Gherardi, N. Atmospheric Pressure Low Temperature Direct Plasma Technology: Status and Challenges for Thin Film Deposition. Plasma Processes Polym. 2012, 9, 1041−1073. (18) Boscher, N. D.; Vaché, V.; Carminati, P.; Grysan, P.; Choquet, P. A simple and scalable approach towards the preparation of superhydrophobic surfaces − importance of the surface roughness skewness. J. Mater. Chem. A 2014, 2, 5744−5750. (19) Moreno-Couranjou, M.; Manakhov, A.; Boscher, N. D.; Pireaux, J. J.; Choquet, P. A Novel Dry Chemical Path Way for Diene and Dienophile Surface Functionalization toward Thermally Responsive Metal−Polymer Adhesion. ACS Appl. Mater. Interfaces 2013, 5, 8446− 8456. (20) Manakhov, A.; Moreno-Couranjou, M.; Boscher, N. D.; Rogé, V.; Choquet, P.; Pireaux, J. J. Atmospheric Pressure Pulsed Plasma Copolymerisation of Maleic Anhydride and Vinyltrimethoxysilane: Influence of Electrical Parameters on Chemistry, Morphology and Deposition Rate of the Coatings. Plasma Processes Polym. 2012, 9, 435−445. (21) Boscher, N. D.; Duday, D.; Verdier, S.; Choquet, P. Single-Step Process for the Deposition of High Water Contact Angle and High Water Sliding Angle Surfaces by Atmospheric Pressure Dielectric Barrier Discharge. ACS Appl. Mater. Interfaces 2013, 5, 1053−1060. (22) Mauchauffé, R.; Moreno-Couranjou, M.; Boscher, N. D.; Van De Weerdt, C.; Duwez, A.-S.; Choquet, P. Robust bio-inspired antibacterial surfaces based on the covalent binding of peptides on functional atmospheric plasma thin films. J. Mater. Chem. B 2014, 2, 5168−5177. (23) Friedrich, J. The Plasma Chemistry of Polymer Surfaces: Advanced Techniques for Surface Design; Wiley-VCH: 2012; p 11. (24) Yasuda, H.; Hsu, T. Some Aspects of Plasma Polymerization Investigated by Pulsed R.F. Discharge. J. Polym. Sci., Polym. Chem. Ed. 1977, 15, 81−97. (25) Friedrich, J. Mechanisms of Plasma Polymerization − Reviewed from a Chemical Point of View. Plasma Processes Polym. 2011, 8, 783− 802. (26) Klages, C.-P.; Höpfner, K.; Thyen, R. Surface Functionalization at Atmospheric Pressure by DBD-Based Pulsed Plasma Polymerization. Plasmas Polym. 2000, 5, 79−89. (27) Camporeale, G.; Moreno-Couranjou, M.; Bonot, S.; Mauchauffé, R.; Boscher, N. D.; Bebrone, C.; Van de Weerdt, C.; Cauchie, H. M.; Favia, P.; Choquet, P. Atmospheric-Pressure Plasma Deposited Epoxy-Rich Thin Films as Platforms for Biomolecule Immobilization − Application for Anti-Biofouling and XenobioticDegrading Surfaces. Plasma Processes Polym. 2015, 12, 1208−1219. (28) Coulson, S. R.; Woodward, I. S.; Badyal, J. P. S.; Brewer, S. A.; Willis, C. Ultralow Surface Energy Plasma Polymer Films. Chem. Mater. 2000, 12, 2031−2038. (29) Rinsch, C. L.; Chen, X.; Panchalingam, V.; Eberhart, R. C.; Wang, J.-H.; Timmons, R. B. Pulsed Radio Frequency Plasma Polymerization of Allyl Alcohol: Controlled Deposition of Surface Hydroxyl Groups. Langmuir 1996, 12, 2995−3002. (30) Detomaso, L.; Gristina, R.; Senesi, G. S.; d’Agostino, R.; Favia, P. Stable plasma-deposited acrylic acid surfaces for cell culture applications. Biomaterials 2005, 26, 3831−3838. (31) Boscher, N. D.; Hilt, F.; Duday, D.; Frache, G.; Fouquet, T.; Choquet, P. Atmospheric Pressure Plasma Initiated Chemical Vapor Deposition Using Ultra-Short Square Pulse Dielectric Barrier Discharge. Plasma Processes Polym. 2015, 12, 66−74. (32) Hilt, F.; Boscher, N. D.; Duday, D.; Desbenoit, N.; LevaloisGrützmacher, J.; Choquet, P. Atmospheric Pressure Plasma-Initiated Chemical Vapor Deposition (AP-PiCVD) of Poly(diethylallylphosphate) Coating: A Char-Forming Protective Coating for Cellulosic Textile. ACS Appl. Mater. Interfaces 2014, 6, 18418− 18422. (33) Boscher, N. D.; Choquet, P.; Duday, D.; Verdier, S. Advantages of a Pulsed Electrical Excitation Mode on the Corrosion Performance of Organosilicon Thin Films Deposited on Aluminium Foil by L

DOI: 10.1021/acs.macromol.7b00461 Macromolecules XXXX, XXX, XXX−XXX