Real-time optical reflection of incident p-polarized light near Brewster’s angle shows that after drop-casting iron oxide nanoparticles (NPs) in heptane on top of a diethylene glycol (DEG) liquid substrate, an iron oxide NP layer forms at the DEG/heptane interface, and it self-limits to a monolayer even when there are excess NPs dispersed in the upper heptane phase. Most modes of NP self-assembly do not self-limit growth after the formation of a single monolayer. Observations are compared to a reflection model incorporating the reflectances expected at each interface. An effective medium model of the dielectric constant is used to model the reflectance of the NP layer at the DEG/heptane interface.

To form these NP assemblies, 2 mL of DEG were first deposited into a glass Petri dish (diameter = 3.45 cm) to form a ∼2.14 mm-thick liquid substrate (if the surface were flat). A p-polarized He-Ne laser (632.8 nm, plasma filter) was expanded from below to form a 2.21 cm × 1.32 cm elliptical beam at the center of the DEG/air interface, to average over this region, and all reflected beams other than those from the glass bottom of the Petri dish were imaged on a photodiode, on which they overlap. The laser was adjusted to hit the DEG/air interface at Brewster’s angle (34.65°) to minimize reflectance from this interface. Then, 1400 μL of heptane were deposited on the DEG substrate to form a ∼1.50 mm-thick upper layer reservoir and, finally, 120 μL of a heptane colloid of NPs were drop-cast on top. This quickly mixed with the heptane reservoir to form a more dilute, ∼1.63 mm-thick heptane colloid layer. The Petri dish was then capped to slow the heptane evaporation. The NPs were spherical iron oxide NPs (Fe₂O₃, 11.8 nm core diameter, capped by oleate) synthesized by standard methods. The number of NPs in the heptane colloid drop was characterized by the number of close-packed NP MLs that would be expected to form on the DEG substrate after heptane evaporation, averaged over the surface; this is termed the number of ML-equivalents being drop-cast.

The light beams reflected from the DEG/heptane colloid and heptane colloid/air interfaces were monitored by a photodiode for 240 min, way before the heptane upper layer evaporated. Before the heptane was added, the reflected fraction from the DEG/air interface was very small, but not zero, because of the surface and laser beam curvatures. After it was added and before the NPs were drop-cast, the reflectances from the DEG/heptane and heptane/air interfaces were still very small because the refractive indices of pure DEG and heptane are nearly the same (below). After the NPs were added and a NP layer formed at the DEG/heptane colloid interface, the reflectance from that interface was expected to initially increase monotonically and linearly with layer thickness at the interface (for up to ∼6 MLs, corresponding to ∼ a quarter wavelength in this medium).