This selection of images drawn from Felice Frankel's book, On the Surface of Things, invites us to explore the visual language of science where form, pattern and material behavior, converge at the boundary between observation and abstraction. Frankel's photographs are not simply documentation -- they are provocations. By transforming scientific processes and phenomena into striking visual compositions she opens a space where art and science reflect one another. Through her lens, surface tension crystallization and microscale structures become a vocabulary of texture, rhythm, and light. As one of the leading figures in the field of scientific image making, Frankel, who began at MIT in 1994, is a research scientist in the department of Chemical Engineering with support from Materials Science and Engineering and Mechanical Engineering. She challenges us to see beyond data, to recognize the aesthetics of inquiry, and to appreciate the narrative potential of even the smallest material detail.
Felice (www.felicefrankel.com) is a Fellow of the American Association for the Advancement of Science and was awarded a Guggenheim Fellowship. She was previously a Senior Research Fellow in Harvard University’s Faculty of Arts and Sciences in the Initiative for Innovative Computing (IIC), and a Visiting Scholar at Harvard Medical School’s Department of Systems Biology.
Working in collaboration with scientists and engineers, her images have appeared in outlets such as Nature, Science, JACS, PNAS, Langmuir, Joule, National Geographic, Newsweek, Scientific American, Discover, Popular Science, and New Scientist, among others. She is the author of many books. Her most recent book, Phenomenal Moments, for young adults, will be published November 4, 2025.
Frankel’s upcoming book, "Phenomenal Moments: Revealing the Hidden Science Around Us,” published by MITeen, which is MIT’s presses imprint for young adults, is available for pre-order here. Enlisting readers to “be the scientist” through vivid fine-art photographs, Frankel zooms in and out on beautiful and brilliant moments all around us to reveal the chemical, natural, or physical processes—from viscosity and venation to chlorophyll and capillary action—behind scientific phenomena.
1. Retroreflectors
William Bernett, formerly at 3M, supplied me with a sample of the material. I photographed it using a combination of reflected and transmitted light.
This array of corner cubes is a 3M product and is part of a broad technology developed by this and other companies based on microreplication. This material is used for optical retroreflectors. Others are used to enhance the brightness of computer displays by directing more of the light toward the user, to distribute light in illumination systems, and to produce abrasives in which the shapes and orientations of the surfaces used in cutting are precisely controlled.
2. Oil Slick
After hosing down my driveway on a gray Sunday afternoon I searched for one of the more interesting puddles and dropped a bit of oil onto it. I waited half an hour until the diffraction colors became interesting. With the camera and 105mm lens on my tripod I shot perpendicularly to the slick.
3. Oxide Layer on a Copper Pan
A friend let me borrow this magnificent au gratin dish. I shot its oxidized underside using daylight with the 105mm lens.
4. Silicon, Etched by Light
Theodore Bloomstein’s work for his doctoral thesis in the Department of Electrical Engineering and Computer Science at MIT involved laser etching silicon surfaces. I used his reference numbers as a compositional element.
With this technology, three-dimensional structures can be rapidly machined in silicon to micrometer tolerances. This work benefits the emerging field of microelectromechanical systems (MEMS) by extending milling techniques to new size regimes. MEMS merges electronic and mechanical components to produce sensors and actuators such as pressure transducers, valves, and pumps. Current techniques for fabricating these mechanical components rely on lithographic techniques that limit shapes to extrusions of two-dimensional patterns. Laser processing offers a means to produce mechanical components that are inherently more three-dimensional.
5. Wing of a Morpho Butterfly
I photographed M. sulkowski, from Brazil, in daylight.
6. Computer Monitor Screen
I first digitally colored the black and white portion of an image of my coauthor’s tie (from his home page). I then set up the tripod in front of my computer screen and used a 2x extender on my 105 mm macro and shot with daylight film. I initially tried tungsten but found I preferred daylight film’s color.
7. Stamp for Microprinting
I used tungsten lamps to produce the diffraction colors for postdoctoral fellow James Wilbur’s and Rebecca Jackman’s sample from the Whitesides Lab. A similar image appeared on the cover of Science, August 4, 1995. An elastomeric stamp—made of transparent polydimethylsiloxane (PDMS) is formed by casting the polymer against a master structure that has a pattern of features in bas-relief on its surface. The stamp is “inked” with a compound that forms a self-assembled monolayer onthe surface being patterned: a typical combination is a PDMS stamp, an alkanethiol “ink,” and a flat gold surface. The inked stamp comes into contact with the gold surface; an image complementary to the pattern on the surface of the stamp forms. The resolution of this pattern is remarkable: the roughness of the edges of the pattern is less than 50 nm, even when the process is carried out in the open laboratory. The colors from the front side are those of diffraction of light propagating in air; those from the back side are from diffraction of light propagating in PDMS.
8.Veins of Opal
William Metropolis, Assistant Curator of the Mineralogical Museum at Harvard University, was kind enough to let me borrow a few large samples of opal. I set this particular specimen under daylight and captured the diffracting light with my 55mm micro lens.
9. Inks Bleeding on Fabric
This is a detail of my favorite handmade scarves. The artist had placed the inks on the fabric anticipating how far they would bleed. I shot it under diffused daylight with the 55mm lens. Capillary wicking is almost universal on contact of a liquid with a porous solid, so long as the contact angle of the liquid on the solid is less than 90’. The liquid-vapor interfacial free energy may or may not be important in capillary wicking: if the capillaries are completely closed, wicking may result in very little change in liquid-vapor surface area. In wicking of liquid into this porous fabric, there is be a significant increase in the liquid vapor area, and the surface tension of the liquid is important.
10. Wrinkled Gold
A slab of polydimethylsioxane (PDMS) that had a flat top was placed in an electron-beam evaporator configured to evaporate gold. Heat radiated by the very hot gold caused the surface of the PDMS to warm, and expand. Deposition of gold on the PDMS thus occurred on a thermally expanded surface. When the deposition was completed and the e-beam extinguished, the PDMS cooled and contracted. Since the coefficient of expansion of PDMS is much larger than that of gold, the gold film was put under compression. The stress of this compression was relieved by buckling. Th large gold balls (inadvertently ejected from the surface of the heated gold by too-rapid heating) provided the nucleation points for this buckling.
11. Crystals of Small Particles
I wanted to see how these nano crystals appeared under an optical microscope, Christopher Murray, and Cherie Kagan, then graduate students in Professor Moungi Bawendi’s lab, were usually more interested in scanning electron microscopy (SEM). Here I used a blue filter when making the image which suggested that these crystals show luminescence in the yellow part of the spectrum. This image is an optical micrograph of bordered crystals of colloidal particles of CdSe the surface stabilized by a thin, absorbed film of n-alkanethiolate. The colloidal crystals were produced by depositing a drop of suspension of the colloid in a solution on a fused silica slide, and then allowing the solvent to evaporate under reduced pressure.
12. Small Machines
I photographed the blades of a microrotor fabricated by Chunang-Chia Lin from Professors Stephen Senturia and Martin Schmidt’s lab in the Department of Electrical Engineering and Computer Sciences, MIT. Choosing which plane to keep in focus is always a question in microscopic photography.
13. Migrating Bacteria
The light for this shot was unusual: to read the contours I had to combine front lighting with irregular back lighting. Using a magnifying mirror that reflected an irregular background, I placed the petri dish on an angle. Professor James Shapiro in the Department of Biochemistry and Molecular Biology at the University of Chicago was kind enough to bring me samples from Chicago.
Bacteria are favored organisms for the study of how a stimulus sensed by an organism and converted into a signal inside the cell results in an observable change in behavior. The responses of bacteria to simple chemical signals are reasonably well understood. The behavior of collections of bacteria is much more complicated. The beautiful regularity and complexity of this pattern of Proteus colonies was unexpected. One key to the formation of the visible terraces is the density of growth: when the population of bacteria reaches a certain level, it differentiates a limited subpopulation of swarmer cells that move out and colonize a new region.
14. Ferrofluid
I dropped a few ml of the dark brown liquid on a glass plate and placed that on yellow paper resting on six circular magnets measuring about 1 cm in diameter each. The magnets created the pattern you see.
A ferrofluid is a stable superparamagnetic suspension of a magnetite colloid with sizes of approximately 10 nm—smaller than the domains required for stable ferromagnetism.
15. Moiré Pattern
Draping and overlapping a few loosely woven fabrics and attaching them to my window during the day created some interesting moiré patterns.
16. Nonwetting Surfaces
Rob Nicholson from the Botanic Garden at Smith College, gave me a number of leaves to play with.
17. Molded Plastic Microfabric
A transparent polydimethylsiloxane PDMS) stamp was prepared by casting a liquid pre-polymer across a relief structure of photoresist: the pattern was introduced by conventional photolithography. The PDMS stamp, on contact with a flat piece of glass formed a network of channels with minimum channel diameter of approximately 1 um. A liquid epoxy prepolymer was brought into contact with this network and filled it by capillarity. Thermal or photochemical cross-linking of the prepolymer (using light transmitted through the PDMS) gave a solid network of polymer on the glass, which was then dissolved to release the plastic fabric.
18. Light-Reflecting from a Polymer-Containing Gel
This image is one of a series taken over period of 24 hours. Over time the colors shifted. The image was captured with my digital camera with a 105 mm lens.
A sample of a very high molecular-weight block, co-polymer (poly(styrene-b-isoprene)) was discovered in cumene, and placed in a thin film between a glass slide and a glass cover slip. The colorless polymer phase – separated into a layered structure with layers parallel to the plane of the slide. The separation between layers was in the order of 200 to 300 nm these layers acted as a multi layer dielectric mirror – a little like the oil slick discussed earlier the details of the processes that gave slightly greater layer to layer separations at the periphery of the disc then under its center are not clear, but probably have to do with mass transport coupled to evaporation of solvent from exposed solution at the edges of the glass cover slip.
19. Analysis of DNA
I cut a 4 x 5 piece from Giang Zhou’s original X-ray film from Professor Philip Sharp’s lab in the Department of Biology at MIT.The film recorded DNA sequences. I scanned it on a Leaf 45, and colored it slightly with Photoshop. Tat Stimulatory Factor 1 (Tat-SF1) is a protein produced by a human cDNA clone that encodes this cellular cofactor for human immunodeficiency virus-1 (HIV-1 Tat). The Tat protein stimulates transcription from the HIV-1 long terminal repeat with the help of several cellular proteins, including Tat-SF1. Tat is absolutely essential for replication of HIV-1. This DNA sequencing gel shows the DNA sequence of the NH2-terminal portion of the Tat-SF1 gene.
20. Agate
Using auxilliary tungsten lighting and the stereomicroscope, I photographed a sample of agate from my colleague Elizabeth Connors’ mineral collection.
21. Square Drops of Water
Nickolas Abbott who was then a postdoctoral fellow in the Whitesides lab and now a professor at the University of California at Davis, said we could emphasize the shape of the drops by adding fluorescing dyes to the water without changing the integrity of the science. I used UV as a light source and the image appeared on the cover of Science, September 4, 1992.
The surface on which the drops rest is a hydrophilic, self-assembled monolayer supported on a thin gold film. Micromachining of 1-μm lines into this surface in a square grid exposed bare gold; these lines of bare goldwere covered with a hydrophobic self-assembled monolayer. Drops of water containing dyes were then introduced into the square hydrophilic areas. The drops did not mix between squares, despite the small separation between their edges.
22. Microelectrodes
A similar shot appeared on the cover of Langmuir, October 1995.The work was done in the laboratory of Professor Mark Wrighton, former Provost of MIT, and now Chancellor of Washington University, St. Louis.
23. Surface of Broken Glass
One of my undergraduate photography students at MIT, Katherine Notter, took a wonderful picture of a piece of broken glass and I thought I’d give it a try.
Fracture mechanics is the study of the mechanisms of failure of solids. The wide range of failure mechanisms includes not only cracking but also solids, flowing, slipping along planes in crystal; or generating bubbles by cavitation ahead of the crack tip. Glass is a brittle solid and fails by concoidal fracture in a series of progressive fractures of the siliconoxygen and boron-oxygen bonds that make up the glass, interspersed with pauses while the tension builds to levels that permit the fracture tip to advance. This picture was taken using reflection Nomarski-mode photomicrography, and the colors provide a map of the topography of the surface. Reflection Nomarski mode records lateral gradients in the phase of the light that correspond to gradients in height.