Superhydrophobic Copper Surfaces with Anticorrosion Properties

Dec 15, 2016 - Due to continuous miniaturization and increasing number of electrical components in electronics, copper interconnections have become cr...
1 downloads 9 Views 1MB Size
Subscriber access provided by University of Newcastle, Australia

Article

Superhydrophobic copper surfaces with anti-corrosion properties fabricated by solventless CVD methods Ignasi Vilaró, Jose L. Yagüe, and Salvador Borros ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b12119 • Publication Date (Web): 15 Dec 2016 Downloaded from http://pubs.acs.org on December 19, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Superhydrophobic copper surfaces with anti-corrosion properties fabricated by solventless CVD methods Ignasi Vilaró1, Jose L. Yagüe1* and Salvador Borrós1,2* 1

Grup d’Enginyeria de Materials (GEMAT), IQS-School of Engineering, Ramon Llull University, Via Augusta 390, 08017 Barcelona, Spain 2 Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), 50018 Zaragoza, Spain

ABSTRACT Due to the continuous miniaturization and the increasing number of electrical components in electronics, copper interconnections have become critical for the design of 3D integrated circuits. However, corrosion attack on the copper metal can affect to the electronic performance of the material. Superhydrophobic coatings are a commonly used strategy to prevent this undesired effect. In this work, a solventless two-steps process was developed to fabricate superhydrophobic copper surfaces using chemical vapor deposition (CVD) methods. The superhydrophobic state was achieved through the design of a hierarchical structure, combining micro/nanoscale domains. In the first step, O2 and Ar plasma etchings were performed on the copper substrate to generate microroughness. Afterwards, a conformal copolymer, 1H,1H,2H,2H-perfluorodecyl acrylate - ethylene glycol diacrylate [p(PFDA-co-EGDA)], was deposited on top of the metal via initiated CVD (iCVD) to lower the surface energy of the surface. The copolymer topography exhibited a very characteristic and unique nanoworm-like structure. The combination of the nanofeatures of the polymer with the microroughness of the copper leaded to achieve the superhydrophobic state. AFM, SEM and XPS were used to characterize the evolution in topography and chemical composition during the CVD processes. The modified copper showed water contact angles as high as 163º and hysteresis as low as 1º. The coating withstood exposure to aggressive media for extended periods of time. Tafel analysis was used to compare the corrosion rates between bare and modified copper. Results indicated that iCVD-coated copper corrodes 3 orders of magnitude slower than untreated copper. The surface modification process yielded repeatable and robust superhydrophobic coatings with remarkable anti-corrosion properties.

KEYWORDS: superhydrophobicity, corrosion, plasma etching, iCVD, hierarchical structure, fluoropolymer 1. INTRODUCTION Copper is an engineering material that has found widespread applications due to its excellent thermal and electrical properties. The high electrical conductivity combined with its favorable mechanical properties has enabled a continuous development in electronics and computer technology. In particular, copper is gaining growing importance in the design of new devices in microelectronics. One important application is focused on the architecture of vertical interconnections to form 3D Integrated Circuits (IC).1–4 The constant reduction in size of the components in IC entails the minimization of the dimensions of the electrical interconnections as well. Aluminum has been the most commonly used material to fabricate interconnections in IC. However, there are different problems associated to the use of aluminum when shrinking its dimensions, such as signal propagation delay, increase of the resistive

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

losses or electromigration.5,6 As a consequence, the microelectronic industry is migrating towards the use of copper as the chosen material for the design of these interconnections. Although copper possesses good corrosion properties, exposure to elevated temperatures or high humidity can result in corrosion and thus, in a loss of performance of the device.7,8 Superhydrophobic coatings on metals display remarkable properties against corrosion.9 Superhydrophobicity is achieved by combining rough or textured surfaces with low surface energy materials. Roughening the surface increases the area of a solid leading to higher water contact angles (WCA), what is known as Wenzel state. However, the superhydrophobic behavior is achieved when air can be trapped into the rough surface between the water drop and the substrate, known as Cassie state. The Cassie state shows WCA higher than 150º and contact angle hysteresis (difference between advancing and receding contact angle) lower than 10º. The air trapped acts as a barrier preventing oxidants to attack the metal surface.10 Selfassembling monolayer (SAM) coatings have been widely studied due to their anticorrosive properties.11–14 A dense and ordered monolayer can act as an insulator barrier to protect the surface against corrosion. Typically, alkanethiols and alkyl disulfides have been used to form SAMs on copper.15–17 The major inconvenient for the use of these monolayers arises from their stability. The thiolate bond, formed between copper and the thiol anchoring group, can be oxidized resulting in the detachment of the monolayer from the surface.18 As an alternative, the use of silanebased SAMs has been proposed. Silane molecules can undergo reaction with hydroxyl-terminated surfaces to form covalently bonded monolayers.19 As a consequence, the formation of the covalent bond between the molecule and the substrate permits a more reliable and stable performance. Because copper is normally used in its pristine form, there are no hydroxyl groups available at the surface. Thus, the assembling of silanes on copper surfaces has been rarely investigated. Hoque et al. used diluted nitric acid to oxidize copper and assemble a perfluorosilane monolayer.20 Further research showed that copper exposure to H2O2 solution generates hydroxyl groups that can react with the perfluorosilane to form a siloxy-copper bond and then, reach superhydrophobicity.21 As stated above, the different treatments performed on copper to yield anti-corrosion properties generally involve the use of solvents and/or strong oxidizers. The use of these chemicals can result in unwanted side reactions and damage to the substrate. Conversely, chemical vapor deposition (CVD) techniques allow the surface modification of practically any substrate without the need of using solvents and have proved their potential value in many different fields. Besides, CVD is commonly used to deposit copper for electrical interconnections in IC.22–24 Hence, the deposition of a corrosion protective coating by CVD can be highly desirable to be implemented during the copper step coverage for the design of 3D ICs. In this work, as far as we are concerned, we have developed for the first time a solventless method to fabricate superhydrophobic copper surfaces by plasma etching and a posterior coating process. The first step consists of conducting a plasma etching to induce a hierarchical structure on the metal surface. Instead of using a conventional acid or base etching attack, combined cycles of O2 and Ar plasma are applied to generate micro- and nanoroughness. Afterwards, a copolymer of 1H,1H,2H,2H-perfluorodecyl acrylate and ethylene glycol diacrylate [p(PFDA-co-EGDA)] is deposited on the copper by initiated CVD (iCVD). iCVD is a low energy vapor deposition process to yield polymeric thin films.25,26 An initiator, usually a peroxide, and the monomer(s) are introduced into a vacuum chamber in the vapor phase. The initiator is thermally decomposed into radicals through heated filaments suspended above the stage. These radicals react with the monomer(s) adsorbed on the cooled stage of the reactor via free-radical polymerization mechanism to yield the polymeric film.27 In

ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

addition, iCVD can produce films that conformally coat complex or 3D substrates.28,29 That is, the polymer is deposited along the substrate without altering its geometry. This feature is crucial to maintain the previously plasma-etched modified structure of copper during the coating step and to create the superhydrophobic effect on the surface. In this study, a superhydrophobic copper sample is fabricated by means, first, of plasma etching to generate the required roughness, and then, by deposition of a low surface energy polymer using iCVD. The topography of copper is imaged before and after the etching and the polymerization by AFM and SEM techniques. XPS, XRD and ATR are used to analyze the oxidation state and the chemical composition of the samples. The wetting behavior and anti-corrosion performance of the coating are investigated using WCA measurements and by electrochemical experiments.

2. EXPERIMENTAL PROCEDURE 2.1 Materials and Chemicals Copper coupons (99.95%, 30x30x0.5 mm) were all cut from a copper sheet (500x250 mm, Materials World) using a precision low-speed cut-off saw (Buehler). Nitric acid, hydrochloric acid, sodium hydroxide, sodium chloride, hydrogen peroxide were reagent grade and used as received. 1H,1H,2H,2H-perfluorodecyl acrylate (PFDA 97%, Sigma-Aldrich), ethylene glycol diacrylate (EGDA, PolySciences), 1H,1H,2H,2H-Perfluorodecyltrichlorosilane (PFTS 96%, Alfa Aesar), tert-butyl peroxide (TBPO 98%, Luperox® DI, Sigma-Aldrich) and Cetyltrimethylammonium chloride (CTAC, 25% wt. in H2O solution, Sigma-Aldrich) surfactant were used without further purification. 2.2 Solution phase etching Copper samples were cleaned sequentially with acetone and isopropanol during 5 min each in ultrasounds. Then, copper was treated with 0.1 M hydrochloric acid solution for 5 minutes to remove the surface oxide. Copper samples were submerged in 4 M nitric acid solution with 1.2 mM CTAC surfactant for 20 minutes, cleaned with distilled water and dried with nitrogen obtaining a rough homogeneous copper surface. 2.3 Plasma etching Clean copper samples were put into a custom-built tubular coil plasma reactor, described elsewhere,30 to conduct the plasma etching. A single copper sample is introduced in each experiment always in the same position of the tubular reactor. The process required 2 cycles with 2 different steps. Firstly, samples were treated with oxygen plasma at a working pressure of 0.14 mbar and power of 150 W for 4 minutes. Secondly, the surface was treated with argon plasma at a working pressure of 0.38 mbar and power of 40 W for 2 minutes, causing the etching on the oxide layer. Next a second plasma cycle with the same conditions was applied. Samples were let to cool down at the reactor under argon atmosphere to avoid additional further oxidation. 2.4 Activation and self-assembled monolayer formation

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Solution-etched copper samples were first activated by immersion in hydrogen peroxide aqueous solution 5% (v/v) for 5 minutes and dried with nitrogen. Then, the activated samples were submerged into 1 mM hexane solution of PFTS overnight, enabling the formation of a monolayer. Samples were cleaned by ultrasounds with fresh hexane for 5 minutes, dried with nitrogen and kept in an oven (Memmert, Germany) at 100ºC for 1 hour. 2.5 iCVD polymerization iCVD copolymerizations on plasma-etched copper samples were conducted in a custom-build cylindrical reactor (24 cm diameter and 4-6 cm adjustable height). On top of the reactor was placed a transparent quartz lid, which allows laser interferometry (633 nm He-Ne laser, Thorlabs) for in-situ monitoring of film thickness. PFDA and EGDA monomers were heated at 80ºC and 70ºC respectively and fed into the reactor chamber along with the TBPO initiator, which was fed without heating. In the reaction chamber a Ni/Cr (80/20, Goodfellow) filament was resistively heated at 220ºC, while the stage was kept at 24ºC using a water recirculating system (Julabo). Distance between filament and stage was fixed at 25 mm. Flow rates were set to 0.15 sccm for PFDA, 0.10 sccm for EGDA and 2.0 sccm for TBPO. Working pressure was kept at 0.90 mbar, adjusting the aperture of the pump with a bellow valve. 2.6 Characterization Thickness of iCVD film was evaluated using laser interferometry on a silicon wafer. Films thickness ranged from 80 to 100 nm. Wettability of the superhydrophobic surfaces was evaluated by water contact angle measurements using a Drop Shape Analyser (DSA 100, Krüss). Advancing and receding contact angles were measured on 5 different points using small drops of 5 μL. Wettability measurements were performed before and after any immersion test or electrochemical measurement. Immersion tests were performed submerging copper samples horizontally for 1.5 and 24 h in 4 different media: water, saltwater (NaCl 3.5% w/v), nitric acid aqueous solution (HNO3 1 M) and sodium hydroxide aqueous solution (NaOH 1 M). X-Ray Photoelectron Spectroscopy (XPS) experiments were performed in a PHI 5500 Multitechnique System (Physical Electronics, United States) with a monochromatic X-ray source (Aluminium Kα line of 1486.6 eV energy and 350 W), placed perpendicular to the analyser axis and calibrated using the 3d5/2 line of Ag with a full width at half maximum (FWHM) of 0.8 eV. The analysed area was a circle of 0.8 mm diameter, and the selected resolution for the spectra was 187.85eV of Pass Energy and 0.8 eV/step for the general spectra and 58.7 eV of Pass Energy and 0.25 eV/step for the spectra of the different elements in the depth profile spectra. In-depth measurements for composition depth profiles were obtained by sputtering the surface with an Ar+ ion source (4 keV energy). All measurements were made in an ultra-high vacuum (UHV) chamber pressure between 5x10-9 and 2x10-8 torr. Fourier Transform Infrared (FT-IR) measurements were performed on a Nicolet iS10 spectrometer in normal transmission mode (Smart Omni Transmission) equipped with a DTGS (deuterated triglycine sulfate) KBr (potassium bromide) detector and KBr beam splitter. The C1s peak was calibrated at 285 eV and used as reference for all binding energies. Attenuated Total Reflectance (ATR) measurements were performed using the Smart iTR module in a Nicolet iS10 spectrometer. Spectra were acquired over the range of 400−4000 cm -1 with a 4 cm-1 resolution for 32 scans. Atomic force microscopy (AFM) images were acquired with a XE-100 (PSIA Inc.) with lateral resolution of 0.15 nm and vertical of 0.05 nm, by non-contact mode. Images

ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

were analyzed by XEP and XEI software for data acquisition and image processing, respectively. SEM images were obtained using a Merlin FE-SEM (Zeiss) with acceleration voltage of 2 kV. Samples with non-conductive surfaces were coated with Pd/Au to avoid charging effects. The adhesion between polymer and substrate was tested using the adhesion test in accordance with ASTM Test Method D 3359. Images of the sample before and after the test were captured using a Leica DM 2500 M microscope. Corrosion resistance was studied using a three-cell electrochemical cell, consisting on an Ag/AgCl reference electrode and a platinum counter electrode. Electrochemical measurements were performed with an Autolab PGSTAT 302N Potentiostat-Galvanostat controller (Metrohm Autolab B.V, Switzerland), using a 3.5 % wt. NaCl solution as electrolyte. Nova 2.0 Software was used for the Tafel plot and corrosion parameters determination. Polarization curves were obtained for unmodified, SAM coated and iCVD coated copper, at a scan rate of 1 mV/s in the range of ±100 mV Vs the open circuit potential (OCP) at room temperature. 3. RESULTS AND DISCUSSION The plasma etching process consisted of two consecutives cycles of O2 and Ar to achieve a hierarchical structure. The experimental conditions (plasma power and time) were optimized for that purpose. The O2 plasma creates a thin layer of oxide, which increases the roughness on the surface. Then, Ar plasma sputtering removes part of this oxide to generate additional roughness, at the same time that limits the oxide layer thickness.31 Overdose of power or time during O2 plasma generates a thick black layer of cupric oxide, which hinders the formation of a hierarchical structure and can compromise the interfacial adhesion between copper and polymer.32 As a consequence, two consecutive plasma cycles were required to obtain a hierarchical geometry with good adhesion properties (see Supplementary Information for further details and images obtained for the different plasma processes and test adhesion). The plasma process not only etches the substrate but also modifies the surface chemistry. In fact, oxygen plasma has been extensively used to change the wettability of materials by generating polar groups on the surface.33–35 The surface chemistry before and after the plasma process was compared by XPS. As observed in the depth profile experiment in Figure 1, the oxygen concentration in the plasmatreated samples is considerably higher than in untreated copper. Pure copper readily oxidizes when exposed to ambient conditions to Cu2O and CuO resulting in a stable oxide layer with a thickness between 1 to 10 nm.36,37 Similarly, the graph for the untreated copper shows the presence of oxygen only at the very top of the surface. After less than one minute of sputtering, there is a sudden decay of the oxygen concentration to practically 0. In contrast, the plasma-treated samples present a deeper content of oxygen due to the enhancement of the oxide layer formation. The oxygen concentration in these samples follows a more continuous and sustained decay. It could be hypothesized that oxygen diffusion through the copper oxide layer thickness is the rate-limiting mechanism during the plasma-oxidation process. Further depth analysis of the oxidation process was done by deconvolution of the O and Cu signals. The high resolution spectra of O 1s and Cu 2p at 3.5, 40 and 140 nm (corresponding to etching times of 30, 360 and 1200 seconds) in depth are presented in Figure 2. The Cu 2p spectra are deconvolved into two peaks based on the existing oxidation states: one at 932.6 eV, corresponding to Cu0 and Cu+, and the second one at 933.2 eV, corresponding to Cu2+. Bibliographic binding energy references for Cu0 and Cu+ are 932.6 and 932.4 eV respectively,38 which makes

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

them undistinguishable for the analysis and thus, both are represented as a unique peak. To determine quantitative and qualitatively the existence of CuO and Cu2O, the oxygen signal was examined by deconvolution of the O 1s signal. The spectra are decomposed into two peaks at 530.2 and 529.6 eV, corresponding to CuO and Cu2O respectively. Moreover, the Cu+/Cu2+ ratio in-depth profile was also studied based on the oxide depth. After decapping, the Cu+/Cu2+ ratio was found to be 0.13, 0.57 and 1.36 at 3.5, 40 and 140 nm respectively. This ratio correlates inversely with the oxygen concentration measured at those depths that was 47, 43 and 11%. This result is in agreement with previous studies indicating that CuO predominates over Cu2O at the very top of the surface,36,39 indicating a more oxidized state close to the interphase. The changes in surface topography and roughness during both modification processes, plasma etching and iCVD deposition, were studied by SEM and AFM. The AFM study helped to optimize the plasma conditions to perform the etching on copper. The results are presented in Figure 3. A typical bare copper topography can be observed in Figure 3a, characterized by its flat geometry with an Sq roughness of just 40 nm. It should be noted that no polishing step of any kind was performed on the sample. Conversely, the copper after the plasma treatment increases its Sq roughness to 430 nm (Figure 3c). The combined cycle of O2 and Ar yields a hierarchical structure with micro- and nano-features due to the oxidation and etching processes occurred on the surface. Images acquired reveal a crease-shape structure along the surface due to the continuous ion bombardment received during the etching. Pal et al40 used a wet chemical etching method to increase surface roughness on copper (Figure 3b). Both methods have displayed similar Sq values, proving the value of plasma etching as an alternative tool to increase roughness on copper surfaces. During the experimental study, Ar and O2 plasma etching were carried out separately to observe the effect on copper. Ar-plasma resulted in a surface uniformly distributed with nanopeaks. In addition, the O2-plasma caused an oxidation of the surface with mountain-like domains in the micrometer range. None of the two processes alone achieved the hierarchical structure required. Nonetheless, consecutives cycles of both of them lead to the appropriate etching of the metal. To obtain the superhydrophobic state is critical to maintain the morphology of the substrate when coating with the fluoropolymer. Deposition of polymers using wet chemistry or PECVD has been proved to not coat 3D objects evenly, which can result in a loss of roughness due to a planarization effect. As explained earlier, iCVD enables conformal coating on complex substrates. In a previous study, Baxamusa and Gleason41 demonstrated that good conformality is achieved when operating at low PM/Psat, that is, the ratio between the partial pressure of the monomer and its saturation pressure. Figure 3d shows the morphology of copper after the p(PFDA-coEGDA) polymerization. The final topography of the coating is comparable to the microstructure observed after the etching process, corroborating the conformal coverage of the copolymer along the surface (Sq = 340 nm). Additionally, FE-SEM images also support the change in the topography induced by plasma and its preservation after the polymer deposition (Figure 4). Moreover, FESEM image at high magnification (Figure 4d) shows the surprising structure adopted by the p(PFDA-co-EGDA). The copolymer reveals a uniform layer with small protuberances arising out in all directions in a worm-like shape. These structures may be originated due to a preferential nucleation-growth induced by a mass transference phenomenon of the PFDA monomer. Additionally, it was observed that the introduction of the EGDA crosslinker is critical for the appearance of these features. However, further research is necessary to validate the hypothesis and this work is currently going on. The diameter of these worms is smaller than 100 nm, which leads to the nanostruturation of the surface. As a result, the iCVD process

ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

maintains the microroughness of the plasma etching, and also, contributes to create nanofeatures. Thus, the combination of two different roughness levels on the surface results in the formation of a hierarchical structure, which is desirable to reach a stable Cassie state. The chemical functionality of the polymer was also studied using ATR-IR spectrometry. Figure 5 presents the spectrum of the deposition of p(PFDA-co-EGDA) on the modified copper substrate, and its corresponding monomers, PFDA and EGDA. Characteristic bands attributed to PFDA were clearly distinguishable in the copolymer. The bands at 1246 and 1207 cm-1 are assigned to asymmetric and symmetric stretching of the -CF2- moiety, and the sharp band at 1153 cm-1 corresponds to the -CF2-CF3 end group.42 However, the EGDA bands in the copolymer are masked due to the high intensity of the fluorinated monomer and cannot be clearly distinguished. Following the polymer characterization, the C, O and F signals were monitored by XPS depth profile (Figure 6). Graph shows that C and F elements are predominant at the top of the surface, and after 1 minute of sputtering the signals practically disappear, as expected for the deposition of a polymeric thin film on copper. The ratio F/C observed on the surface is 1.11 while the theoretical value for the pPFDA homopolymer is 1.31. This difference can be attributed to the incorporation of the EGDA cross-linker into the polymeric chains, which contributes to raise the amount of C, and thus, confirms the incorporation of both monomers into the polymeric structure. Wettability of the surface is a key feature to show anti-corrosion properties. iCVDcoated samples showed WCA higher than 150º supporting the formation of the superhydrophobic state. Furthermore, not only high contact angles are necessary but also low hysteresis to reach superhydrophobicity. In surfaces with low hysteresis, drops scarcely pin on them presenting a slippery behavior.43 Nevertheless, surface reconstruction in fluoropolymers can hinder the slipperiness on the surface. Fluorinated groups when in contact with polar groups, like water, can migrate to the inner part of the polymer increasing the pinning of drops. Coclite et al. demonstrated the superhydrophobic character of iCVD pPFDA films controlling the crystallinity of the polymer by tuning the deposition conditions44 or grafting PFDA polymer chains.45 In other studies, we showed that cross-linking of pPFDA polymer chains prevents surface reconstruction by fixing the fluorinated groups on top of the surface.46,47 Table 1 provides WCA and hysteresis values of the surfaces during the different processes conducted. p(PFDA-co-EGDA) iCVD films present superhydrophobicity with a remarkable slippery behavior. We hypothesize that the distribution and separation of the nanoworms found in the polymer facilitate the trapping of air in between these structures when water is placed on top of them, conferring the superhydrophobic effect. Conversely, pPFDA homopolymer results in films with hysteresis larger than 30º, which confirms the effect of the EGDA cross-linker to avoid surface reconstruction. Stability of the polymeric layer was tested by immersion of the substrates in different media: water, a 3.5 M NaCl solution, a HNO3 solution with pH 1, a NaOH solution with pH 12 during 1.5 and 24 h. The WCA and hysteresis were measured before and after exposure (Table 2). Practically no change was observed during the experiments and the superhydrophobicity was maintained, except for the acid and basic media immersed for 1 day. These results suggest that the iCVD coating is robust enough to withstand extreme conditions without a loss of functionality. Finally the anticorrosion performance of the p(PFDA-co-EGDA) films was tested by the measurement of Tafel polarization curves in aqueous chloride media. The iCVDcoated copper was compared to bare copper and to copper modified by selfassembly, since SAMs are commonly used to protect copper against corrosion.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

Copper corrosion due to chloride ions has been extensively studied, and three corrosion mechanisms have been proposed based on the different species generated during the oxidation process:48  + 2 ↔  +

(1)

 ↔  +  + 2 ↔ 

(2)

 +  ↔  +  +  ↔ 

(3)

The Tafel polarization curves for bare copper, copper coated with a fluorosilane SAM and copper coated by iCVD with a fluoropolymer in a 3.5% NaCl solution fresh prepared are illustrated in Figure 7. The corrosion potential, corrosion current density, the corrosion rate and the corrosion inhibition efficiency (ηi) are summarized in Table 3. In comparison to the unmodified copper, the corrosion current decreased considerably for both the SAM and iCVD samples. However, for the iCVD sample the corrosion current is two orders of magnitude lower than for the SAM. The protection ability of both coatings was studied using the corrosion inhibition efficiency, which can be calculated using the following equation:  (%) =

   −  ∗ 100  

(4)

being i0corr and i’corr the corrosion current density for the bare copper and for the coated copper respectively. Having a similar WCA both coatings, the higher efficiency for the iCVD coating shows a denser and pinhole free layer than the one formed with the SAM, and thus, a better protection. The polymeric layer is thick enough to enhance the blocking effect compared to the monolayer, and thin enough to preserve the hierarchical structure of the surface. In addition, the air trapped in the hierarchical structure diminishes the contact of water with the substrate interphase. As a result, the chloride ions have more difficulties to reach the copper surface and corrode the metal. Also, the corrosion potential shifted to more noble values, indicating the enhanced corrosion properties of the coating. In contrast, all the experiments for the SAMs resulted in a corrosion potential slightly lower than the unmodified copper. This effect can be explained due to appearance of defects in the monolayer.49 Finally, the corrosion rate for the unmodified copper was calculated to be 51 µm/year, 2.5 µm/year for the SAM and 0.077 µm/year for the iCVD coated copper, demonstrating the potential use of the iCVD films as anti-corrosive coatings. 4. CONCLUSIONS AND FUTURE WORK In conclusion, a novel solventless approach to produce superhydrophobic copper surfaces has been proposed in this study. To achieve the superhydrophobic state, the copper surface was etched by alternating O2 and Ar plasma to yield a hierarchical structure, followed by the deposition of a conformal fluoropolymer layer via iCVD. AFM, SEM confirmed the micro- and nano-roughness induced by the plasma etching, and the preservation of the hierarchical topography due to the conformal iCVD polymerization. Furthermore, the iCVD copolymer presented distinctive nanowormlike features, which complements the nanostructuration of the surface. The modified surfaces showed superhydrophobicity (WCA>160º) with extraordinary low hysteresis (