From the Nanoscale Device Laboratory (aka 'nanolab'), under the direction of Dr. Sumita Pennathur in the Mechanical Engineering Department at the University of California, Santa Barbara in Santa Barbara, CA. The research in the Nanoscale Device Laboratory is focused on novel studies of chemical and biological species using fabricated nanoscale devices. The scope of the research program is broad, spanning the fields of Physics, Biology, Chemistry, and Engineering. The research goals are just as expansive, focusing on the fundamental science of nanoscale systems, while also exploring exciting technological possibilities. People working on this project: Justin Massey The workhorse technique for microfluidic device fabrication is standard semiconductor device processing, which includes time-consuming and expensive trips to a semiconductor clean room for lithography, etching, and deposition on silicon and silicon dioxide surfaces. Therefore, in the last few years there have been many advances towards the development of rapid prototyping techniques on plastic or borosilicate glass. More recently, researchers have printed microfluidic mould patterns onto pre-stressed thermoplastic sheets of polystyrene (e.g. Shrinky Dinks®).The printed patterns are then used to create channels for fluidic manipulation2. The technique allows for a microfluidic mould to be completed with only a LaserJet printer and toaster oven within minutes. When Shrinky Dinks® are heated they shrink to their original size, also shrinking laser ink on the substrate. The pre-heated size of the Shrinky Dink® is approximately sixty five percent larger than after it shrinks. Therefore, features printed with laser ink can create micron-scale patterns. Polydimethylsiloxane (PDMS) polymer can be poured over this mould and hardened to create open microfluidic channels. The hardened PDMS can then be annealed or press-fitted to a glass slide to create a sealed microfluidic device. Inlet and outlet holes can be cut into the PDMS using a biopsy punch. The fabrication process and device are shown in Figure 1. Though this process is much simpler and less time consuming than traditional microfluidic device fabrication, it is also very inconsistent. Past devices created from this process have all been different in characteristics including roughness of surface, variations in final dimensions due to non-uniform shrinking and ease of adhesion to the glass slide. There are many factors that can change the properties of each device including the use of different printers or copiers, different types of PDMS and the variation of curing times. Different printers can change the characteristics of a device, such as smoothness and precision. Some printers may not create a mould that is well defined to the naked eye, but the channel may be very smooth and more beneficial to certain experiments. Other printers create channels that are closely related to the virtual copy, but the roughness properties may not be desirable. The reduction of roughness in the PDMS is important for fluid flow experiments as well in the adhesion to the glass slide. PDMS curing times will also change the properties of fluid flow with each device, even with the same mould. If the PDMS is cured too long, it will lose its natural adhesion properties due to evaporation and will never stick to the glass slide. If it is not cured long enough, then the particles from the PDMS can still contaminate the channels and ruin potential experiments. For the experimental purposes of this channel it is most important to keep it as clean as possible. There are also problems with delamination of the PDMS due to uneven surface contact with the glass slide. Figure 2: Delamination of PDMS at inlets from uneven bonding surface In Figure 2, the delamination of the PDMS from the slide causes the blue dye to leak out of the channel. The mould not being uniformly flat caused this delamination. Using the most resolute printer setting on the LaserJet Lab Printer, we were able to create six samples with the starting channel width of .500 millimeters, length of 32.33 millimeters and depth of .06 millimeters. The average post shrunken channel dimensions have a width of .120 millimeters, length of 14.6 millimeters and a channel depth of .037 millimeters. Data from measuring the post heated channels indicates that the Shrinky Dink® does not shrink the same rate in all directions. This non-uniform shrinking may be from the Shrinky Dink® manufacturing aimed at being a toy rather than a scientific device. This inconsistency causes problems with creating multiple moulds with the same print. We are currently characterizing and improving this rapid-prototyping technique by optimizing the following components and instruments: the LaserJet ink and printer, the oven used to heat the mold, the manufacturer of the pre-stressed polystyrene sheets, the type of PDMS and curing agent, and the instrumentation for cutting inlet and outlet holes. In order to ensure the flatness of the mould we would like to implement a custom-built metal press to flatten the mould while heating. We would also like to implement the use of a hand-held RF Plasma device to create an optimal bonding surface on the PDMS. After determining the best instrumentation for the prototyping station, we will optimize the protocol for instrument use so that the microfluidic device can be created as accurate and repeatable as possible.We expect advances in this rapid-prototyping technique so that these devices can soon be used for simple undergraduate microfluidic experiments or to create a quick proof of concepts for graduate research. REFERENCES: [1] C.S. Chen, D.N. Breslauer, J.I. Luna, A. Grimes, W.C. Chin, L.P. Lee, and M. Khine; Shrinky-Dink microfluidics: 3D polystyrene chips Lab on a Chip 8, 4, 622-624 (2008). Text from: http://engineering.ucsb.edu/~nanolab/
Shrinky Dinks
