Miniature droplet lenses for mobile microscopy

The flawless surfaces of liquid droplets have long captivated optical scientists and engineers, who have sought to manipulate them for microlasers, optical resonators, and microlenses. Droplets are made of molecules held together by intermolecular cohesive forces, which work along the interface of liquid and air¹ to create a smooth surface with a discrete shape. Rays of light propagating through transparent droplets can take many pathways, such as reflection, refraction, dispersion, or total internal reflection.

Our work focuses on using droplets to manufacture miniature lenses, which are increasingly required for mobile devices, such as laptops, LEDs, and smartphones. To date, conventional polymer injection lens-molding methods have met this demand. However, prototyping and equipment tooling of these processes comes at high cost. We propose an easy and inexpensive method of harvesting solid lenses of varying focal lengths just by hanging and curing droplets of different volumes.

Surface tension maintains the shape of a droplet, but external influences (heat, gravity, and pressure) may destabilize its shape. It is possible to use physical or chemical mechanisms, such as fluidic pressure, surface tension, hydrophobicity, evaporation, or electrowetting to manipulate and shape the droplet. Recent experiments show droplets comprising two miscible liquids with varying rates of evaporation are engaged in a dynamic interplay of self-propelled motions.² Aerosol science exploits the optical properties (scattering and refraction) of micro and macro droplets to study dynamic transient events in combustion and pollutants. However, because it is difficult to maintain or control the shape of larger droplets, these are ill-suited to fabricating high-quality miniature (mm) optical components. We can use tunable liquid lenses, encasing charged droplets within transparent flexible silicone holders, but the complexity in actuating these lenses makes them unsuitable for day-to-day use. Moreover, the simplicity and quality of conventional optics still hold considerable advantages. In refractive lenses, focal length is intricately linked to curvature, which we can retain by freezing a drop. A frozen water droplet may be subject to thermal expansion and shape distortion. However, when we expose to light a mixture of photoinitiators and high-index liquid polymers, we can freeze the droplet instantly, with minimal changes in shape. One example is the use of UV curing to fabricate polymer lenses, which show excellent optical properties.^{3, 4} Grilli⁵ and Wilbur⁶ and their coworkers found ways to freeze transparent viscoelastic liquids, or elastomers (liquid pre-cured polydimethylsiloxane, or PDMS), which solidify at low temperatures (<200°C).

To learn how to effectively harness droplets, we turn to commercial technologies such as inkjet and 3D printers, nebulizers, and cell sorters. These applications exploit droplet separation and coalescence for printing, aerosol delivery, and encapsulation of biological cells. In 3D printing, polymer microdroplets ejected from high-temperature nozzles build physical models, with the advantage that there is little material loss by comparison with subtractive manufacturing. Optics developers use 3D printing with photocurable clear polymers and specialized printing units to increase each layer and build up good quality optics for lighting. However, these have reduced optical transparency and none of the antireflection properties of conventional lenses. To address these issues, we used low-temperature-cured PDMS, which has excellent optical transparency, high resistance to heat (no browning at >200°C), and a low refractive index (∼ 1.41).

We used an ‘inverted’ additive fabrication process, adding smaller layers of drops to a previously cured droplet surface to increase curvature and the resolving power of the lens.⁷ A single hanging droplet forms a parabolic profile of a positive refractive lens, offering one focal length. The technique enables production of a range of lenses with different resolving powers, with each lens resembling an ice drop (see Figure 1). Forming PDMS lenses with different numbers of drops—see Figure 2 (top)—enabled microscopic imaging resolution (4μm) with miniature cameras—see Figure 2 (middle)—and control of light divergence from an LED: see Figure 2 (bottom). We incorporated the lenses into a flexible, low-cost (∼$2) microscope attachment (see Figure 3) that works almost as well as a clinical dermascope using a polymer lens, making it particularly suitable for in-field diagnostic applications.

Figure 1. Left: A frozen ice drop. Right: A droplet-shaped lens suspended from a fingertip.

Figure 2. Top: Four droplet lenses with varying curvature and focal lengths. Middle: Different magnification and imaging resolutions of the droplet lenses using a miniature imaging sensor. Bottom: Controlling light divergence with the droplet lenses.

Figure 3. Comparing the performance of the new elastomer lens with a standard polymer lens. Top: Light microscopy resolution. Bottom: Dermatoscopy of the common mole. (Image courtesy of Kar Gay Lim, Macquarie Health.)

Fabrication of the lenses requires only an oven, a glass microscope slide, and PDMS. First, we drop a small amount of PDMS onto the slide and bake it at 70°C for 15 minutes to harden, creating a base. Then, we drop another layer of PDMS onto the base and flip the slide over. Gravity pulls the new droplet down into a parabolic shape. We bake the droplet again to solidify the lens, after which we can add more drops as needed to hone the shape and increase the lens imaging quality.

The hanging droplet fabrication resulted in a lens surface roughness of only 5–10nm, which is more than an order of magnitude lower than for a conventional polished lens. However, asphericity of the droplet lenses reduces with increased layering, which results in prominent pincushion distortion. To overcome this, we are now developing lenses shaped by gravity that have high asphericity and resolving power to significantly reduce the pincushion effect in the field of view.

The author acknowledges financial support from the Discovery Translation Fund and innovation grants from Plant Biosecurity CRC.