Assembly of Colloidal Particles by Evaporation on Surfaces with Patterned Hydrophobicity
Original entry: Lidiya Mishchenko, APPHY 226, Spring 2009
Fengui Fan, et. al., Langmuir, 2004, 20(8)
"Drops containing suspended particles are placed on surfaces of patterned wettability created using soft lithography; the drop diameter is large compared to the dimensions of the patterns on the substrate. As the three-phase contact line of the drop recedes, spontaneous dewetting of the hydrophobic domains and flow into the hydrophilic domains create discrete fluid elements with peripheries that can mimic the underlying surface topography. Suspended particles are carried with the fluid into the wetted regions and deposit there as the discrete fluid domains evaporate. If particle volume fractions are sufficiently high, the entire wetted domain can be covered with colloidal crystals. At lower volume fractions, flow within the evaporating fluid element can direct the deposition of colloidal particles at the peripheries of the domains. High-resolution arrays of particles were obtained with a variety of features depending upon the relative size of the wetting regions to the particles. When the wetting region is larger than the particles, three-dimensional and two-dimensional arrays of ordered particles mimicking the shape of the wetting pattern form, depending on the particle volume fraction. For lower volume fractions, one-dimensional (1-D) arrays along the wet/non-wet boundaries form. When the particle size is similar to the height of fluid on the wetted domain, zero-dimensional distributions of single particles centered in the wet regions can form for wetted squares or 1-D distributions (stripes) form along the axis of striped domains. Finally, when the wetting region is smaller than the particle size, the particles do not deposit within the features but are drawn backward with the receding drop. These results indicate that evaporation on surfaces of patterned wetting provides a highly parallelizable means of tailoring the geometry of particle distributions to create patterned media."
Soft Matter Example:
The idea behind this paper is straightforward. We already know that colloids can assemble in evaporating droplets through capillary bridges. Also, we know that with pinned evaporating droplets, there is mass transport to the evaporating contact line (contrinuty equation). Thus, can we force particles to assemble in only certain regions using evaporation?
Fan et. al. simply used the concept of contact angle hysteresis and wetting of defects to preferrentially deposit colloids in certain places. By patterning hydrophillic "defects" on a hydrophobic substrate and placing a large drop with suspended colloids on this substrate, they relied on dewetting phenomena (and pinning at the wetting-nonwetting regions) to leave small areas of fluid behind (on the defects) from an evaporating droplet. Their results are summarized in the figure.
There are several scaling laws involved in this method which allowed for greater control of this method. In order to allow for particle deposition, the speed of convection of the particles to the interface needs to be faster than the time required for the droplet to dewet from the defect. If there are no particles at the dewetting interface, there will be no deposition. However, this is probably always the case since the speed of convection of the particles is probably very close to the convection of the fluid, and the fluid clearly evaporates slower than the dewetting process.
In order to have particles assemble in an ordered fashion (which doesn't seem to be the case in this paper judging from the SEMs in the figure), the evaporation of the fluid needs to be slow enough to allow the particles to reach some sort of low energy confidguration (fcc). This can be achieved with lowering the ambient temperature or raising the humidity. Also, if the concentration of the particles is low, they will most likely pack at the pinned edges of the evaporating drop. When colloids are far apart, they do not form capillary bridges between them (which assemble them together at higher concentrations).
Finally, a scaling law the paper did address was how the particle diameter compared to the height of the water left behind at the defect. If the colloid diameter is larger than the height of the remaining fluid (as determined by the receding angle of the contact line), the colloid will not be deposited on the defect (but will be drawn back into the "parent" droplet). The energy of the defect is probably not large enough to oppose the pull of the colloid back into the parent drop (due to its large mass, and the large energy required to create an air-colloid interface).