Wetting in Color: Colorimetric Differentiation of Organic Liquids with High Selectivity
[Ian Burgess, Spring 2012]
Every liquid has a surface tension. Likewise, every liquid will display a contact angle on an ideal flat surface with a given surface chemistry (where this contact angle could be 0). Simple diagnostics that differentiate liquids based on surface tension and wettability have the potential therefore to be useful in a broad range of scenarios. This paper presents a form of colorimetry based on wetting, where organic liquids induce the appearance of color patterns in an indicator material that are highly sensitive to the liquids' wetting properties.
Central to the operation of this device is the porous inverse-opal films (see B. Hatton et al, Proc. Nat. Acad. Sci. USA 107, 10354-10359 (2010)) that serve as its structural basis. The highly regular porosity of this structure serves two purposes: i) it is responsible for the film's iridescent color (when the pores are air-filled) that disappears when the pores are filled with liquid (due to refractive-index matching); and ii) it facilitates the highly selective wetting that this device exploits.
The paper starts by discussing the correlation between the symmetry of the structure and the selectivity of wetting expected. The pore structure consists of a face-centered cubic lattice of spherical air pores in a silica monolith. Each pore is connected by roughly circular openings, called "necks" in the paper. Due to the discontinuous and re-entrant curvature at the necks, liquid fronts show a propensity to pin at these necks, preventing the wetting of the structure. Sufficiently wetting liquids (sufficiently low intrinsic contact angle) will be able to overcome pinning and penetrate the structure. As shown in Fig. 1, the smaller necks, the higher propensity for pinning. The size of a neck defines a critical intrinsic contact angle for fluid penetration. When an inverse opal has a distribution of neck sizes, the penetration of the structure will occur through the necks that are smaller than the critical value defined by the contact angle, but will get pinned at the others. The filling in this regime is well described by a bond percolation with a fixed connectivity. Using neck distributions derived from SEM images, it is estimated that the response should be sensitive to contact angle differences of less than 5 degrees.
To tailor the sensitivity of the response to differentiate two specific liquids, the surface chemistry must be tuned such that the contact angle of one liquid is above the infiltration threshold, but below threshold for the other liquid. For a sensor, these liquids should be chosen by the application (i.e. they can be anything) and so the surface chemistry must be able to be tailored on a continuous scale. To accomplish this, the authors develop two different approaches. The first shown in Fig. 2 below (A,B) consists of laterally patterning locally homogeneous, but mixed monolayers derived from a mixture of two alkylchlorosilanes with different pendant groups. The second, shown in C,D entails vertically grading the surface chemistry such that the contact angle is increasing with depth. Both of these approaches allow different liquids to produce mutually distinct color patterns (shown in schematics in B, D).
Translating wettability into a colorimetric signature of a liquid could lead to a broad class of simple sensors for a variety of liquid-authentication or liquid-identification applications because surface tension and contact angle are quantities that can be defined for every liquid has a surface tension. To illustrate the flexibility of this device the authors show four examples in the last figures of the paper where different closely related pure substances (methanol, ethanol, isopropanol; hexane, heptane, octane, nonane and decane), simple mixtures with different proportions (resolving a 2.5% water dilution change in ethanol) and different complex mixtures (samples of gasoline and diesel) are differentiated colorimetrically. These examples show how principles of wetting and pinning can be exploited to make useful devices. One real challenge for this technology going forward will be to impart chemical specificity. There will always be an infinite number of liquid mixtures that can produce the same contact angle on a given surface and so a sensor built off of such a general property of liquids will not be effective unless it is differentiating liquids from a pre-known finite list of possible unknowns. More generally (i.e. beyond wetting-based devices), there is a tradeoff between specificity and generality in sensor design. Sensors that are highly specific (i.e. lock-and-key) need to be redesigned for every new target (poor generality), however generalizable sensors that operate in response to properties as general as wetting can work in many environments but are difficult to make specific because of the large class of substances that can change the response.