Article pubs.acs.org/Langmuir
Impact of Nanoscale Surface Heterogeneity on Precursor Film Growth and Macroscopic Spreading of [Rmim][NTf2] Ionic Liquids on Mica Zhantao Wang and Craig Priest* Ian Wark Research Institute, University of South Australia, Mawson Lakes, SA 5095, Australia S Supporting Information *
ABSTRACT: The connection between the interfacial properties of ionic liquids and their wetting behavior has been studied very little to date and not at all on heterogeneous surfaces. Therefore, we have investigated the static and dynamic wetting for a family of ionic liquids, 1-alkyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, [Rmim][NTf2], on mica, where R represents an ethyl, butyl, or hexyl alkyl chain on the imidazolium ring. Spreading is impacted greatly by a precursor film that forms on both homogeneous and heterogeneous mica surfaces. Macroscopically, the initial viscous spreading of the ionic liquid droplet on bare mica occurs within seconds but is then followed by a very slow relaxation that can be closely correlated with the typical time-scales of the precursor film growth. The contact angle for [emim][NTf2] and [bmim][NTf2] relaxes from about 40° to 23° over 30 and 90 min, respectively. For [hmim][NTf2], the process takes approximately 24 h and approaches complete wetting. The thickness of the precursor films for [emim][NTf2], [bmim][NTf2], and [hmim][NTf2] were 0.53, 0.65, and 1.0 nm, respectively, according to atomic force microscopy (AFM). These values are consistent with a monolayer of ionic liquid cations on mica, rather than ion pairs. A monolayer of octadecylphosphonic acid (OPA) on mica prevents both the formation of a precursor film and the relaxation of the contact angle. However, only a partial surface coverage of ∼60% OPA is required to have the same effect. Quenching of precursor film formation (and associated contact angle relaxation) is due to an increasingly connected network of OPA regions that closes the nanoscale paths of bare mica on which the precursor film can develop via surface diffusion.
■
and thin films of IL have been observed at the perimeter of IL droplets that were deposited on mica from evaporating methanol solutions.13,15 Surface forces apparatus (SFA) and AFM studies have shown that these layers can exist as a solidlike interfacial phase depending on the solid surface present13,15 and can exhibit enhanced viscosity.19 Displacement of these layers by an AFM tip requires forces from a few nanonewtons10−12,14,17,22 to several tens of nanonewtons,29 which is a good indication of the strength of the interactions involved. Although studies employing AFM, SFA, and other techniques have correlated the thickness of these layers to the molecular volume of the IL, differences in the reported values exist. For [emim][NTf2] and [bmim][NTf2], the reported values of the layer thickness vary between 0.4 and 0.75 nm12,14,17,18 and between 0.5 and 0.9 nm,13,19,20 respectively. Alkyl chain length has also been shown to affect the IL orientation at the surface11,30 and thickness of solvation layers.16 Furthermore, carbonaceous contamination from air at cleaved mica surfaces can alter the morphology of thin IL films grown in ultrahigh vacuum.31 While these interfacial effects are of great interest
INTRODUCTION Due to their unique properties, ionic liquids (ILs) have attracted immense interest from science and engineering communities. Of particular interest is their negligible volatility and the plethora of possible ion-pair combinations that offer an unparalleled ability to tune both physical and chemical properties of ILs for specific applications. In the vast majority of cases, understanding the interfacial behavior of ionic liquids will be essential for their successful application, e.g., as ‘green’ solvents,1,2 lubricants,3 energy storage media4 and solar cells,5−7 optics,8 and in nonaqueous electrochemistry.9 Of interest here is IL wetting, which is of fundamental importance wherever an IL comes into contact with a solid surface and may dictate both spreading kinetics and stability of IL coatings or films. For the first time for ILs, we present evidence for a complex interplay between macroscopic wetting, nanoscale prewetting (or precursor films), and nanoscale surface heterogeneity, showing that the connectivity of the surface heterogeneity is a critical factor in determining the static and dynamic wetting of these liquids. ILs often form highly organized molecular-scale layers at their interfaces with solid surfaces such as mica, silica, and gold,10−23 and at the interface with air or vacuum.24−27 These layers of ion pairs may extend into the bulk liquid10−14,16,17,19,28 © 2013 American Chemical Society
Received: July 15, 2013 Revised: August 9, 2013 Published: August 13, 2013 11344
dx.doi.org/10.1021/la402668v | Langmuir 2013, 29, 11344−11353
Langmuir
Article
with respect to the envisaged applications of ILs, interfacial ordering of ILs is also likely to impact the statics and dynamics of wetting on a solid surface having adverse or desirable consequences for their end-use. Several key wetting phenomena will become increasingly important in ionic liquid based technologies, including contact angle hysteresis, precursor films, wetting dynamics, and autophobicity, to name a few, with some insightful studies already published, as discussed below. Selected aspects of both static32−35 and dynamic wetting36−38 of ILs have been studied to date, with the majority of publications focusing on static wetting. Restolho et al.33 studied the disjoining pressure isotherm of two ionic liquids, 1-octyl-3methylimidazolium tetrafluoroborate and 1-ethanol-3-methylimidazolium tetrafluoroborate, on alumina, revealing that the dispersive forces are dominant in determining their surface properties. Batchelor et al.32 found that [Rmim]-based ILs wet several solid surfaces similarly to molecular liquids such as water, ethylene glycol, and hexadecane, suggesting that, at least on the surfaces studied, these ILs do not exhibit unusual wetting behavior. Electrowetting has attracted much attention due to the unusual combination of electrical conductivity and organic chemistry afforded by ILs.39,40 Paneru et al. showed that electrowetting using pure and water contaminated [bmim][BF4] in an ambient liquid phase (hexadecane) is consistent with that of molecular liquids, following the YoungLippmann prediction until contact angle saturation is observed at high applied potentials.36 Dynamic wetting, however, has received far less attention. Li et al. studied spontaneous wetting38 and forced wetting (electrowetting)41 of several ILs on a fluoropolymer. The authors demonstrated that the molecular kinetic theory provides a good description of their experimental results when the cubic root of the molecular volume is considered equivalent to the molecular displacement length. There is, however, a great deal more that needs to be understood about the wetting of these liquids for their reliable application in technologies. The published observations of interfacial layering of ILs, including recent observations of thin IL films surrounding droplets deposited from methanol solutions, are suggestive of precursor film formation, which has been observed for conventional liquids such as polydimethylsiloxane.42−44 Precursor films are thin liquid films that spread ahead of the macroscopic droplet front on a solid surface and are known to play an important role in the macroscopic spreading of liquids;45 however, the influence of surface heterogeneity on precursor film formation and growth has been studied very little to date (even for molecular liquids). The negligible volatility of ILs offers a distinct advantage for studying the development of precursor films because the film growth must occur via surface diffusion alone. New evidence for precursor film formation for ILs has emerged very recently;46 however, heterogeneous surfaces have not been considered at all. With no existing study of the formation of IL precursor films on heterogeneous surfaces, this paper addresses the spreading of these unique liquids with respect to the connectivity of the surface heterogeneity. The molecular structure of the anion and cations in the [Rmim][NTf2] ILs studied are shown in Figure 1, where R represents an ethyl, butyl, or hexyl alkyl chain on the imidazolium ring. We exploit the negligible volatility of these ILs and AFM to reveal not only the existence of [Rmim][NTf2] precursor films on mica, but also the relationship between the
Figure 1. Molecular structure of the anion and cations in the ionic liquids used in this study; hydrogen atoms and π-bonds are not shown on the three illustrated cations. Tail-to-tail, a, and tail-to-ring, b, cation dimensions are given for the three cations, estimated using bond lengths and angles in ref 47. Atoms in the illustrated cation structures are color coded; carbon (light blue), nitrogen (dark blue). Cation dimensions, a and b, viscosity, η, and surface tension, γ, for [emim][NTf2], [bmim][NTf2], and [hmim][NTf2] are given below the cation structures.
precursor film, macroscopic spreading, and nanoscale surface heterogeneity. We show that the connectivity of surface heterogeneity is a key factor in precursor film formation and, consequently, the macroscopic static and dynamic wettability. Whenever the bare mica surface is disconnected (i.e., separated by a network of OPA monolayer that partially covers the surface), precursor film growth is arrested due to its dependence on the surface diffusion mechanism, along with the slow contact angle relaxation observed on bare mica surfaces.
■
EXPERIMENTAL SECTION
Three imidazolium-based ILs, [emim][NTf2], [bmim][NTf2], and [hmim][NTf2] (Merck Chemicals, > 99% purity, halides
