[FoRK] Coolest use for Shrinky-Dinks

Ken Meltsner <meltsner at alum.mit.edu> on Wed Dec 5 15:57:32 PST 2007

New researcher without the right fab equipment figures out how to
create microfluidic analysis "chips" using a laser printer and
Shrinky-Dink plastic sheets.  I made a few window decorations when I
was young, lacking the imagination to create such devices:

http://www.rsc.org/delivery/_ArticleLinking/DisplayHTMLArticleforfree.cfm?JournalCode=LC&Year=2008&ManuscriptID=b711622e&Iss=Advance_Article

Now, if I could figure out a use for an old Tootsie roll press and a
Plastigoop molding system....


"Shrinky-Dink microfluidics: rapid generation of deep and rounded patterns

Anthony Grimes(a), David N. Breslauer(b), Maureen Long(a), Jonathan
Pegan(a), Luke P. Lee(b) and Michelle Khine*(a)

(a) School of Engineering, University of California, Merced, USA.
E-mail: mkhine at ucmerced.edu
(b) Department of Bioengineering, University of California, Berkeley, USA
Received 31st July 2007, Accepted 2nd November 2007

First published on the web 20th November 2007

We present a rapid and non-photolithographic approach to microfluidic
pattern generation by leveraging the inherent shrinkage properties of
biaxially oriented polystyrene thermoplastic sheets. This novel
approach yields channels deep enough for mammalian cell assays, with
demonstrated heights up to 80 µm. Moreover, we can consistently and
easily achieve rounded channels, multi-height channels, and channels
as thin as 65 µm in width. Finally, we demonstrate the utility of this
simple microfabrication approach by fabricating a functional gradient
generator. The whole process—from device design conception to working
device—can be completed within minutes.

Introduction

To address the need to create deep and rounded microfluidic channels
without expensive and dedicated tooling, we introduce a novel method
of printing microfluidic channel networks onto commercially available
thermoplastic Shrinky-Dinks in a standard laser-jet printer.
Shrinky-Dinks is a children's toy onto which one can draw a picture
and subsequently shrink it to a small fraction of its original size.
We printed features onto this thermoplastic, and found that after
heating for 3–5 min at 163 °C, printed features shrink isotropically
in plane by approximately 63% from the original printed line width and
length. There is an additional corresponding increase in height of the
features by over 500%. We subsequently used these shrunken features as
a rigid mold for soft lithography.1 The thermoplastic mold is thus
analogous to the commonly-used silicon wafer, which typically requires
photolithographic patterning, for microfluidic applications. Like its
silicon wafer counterpart, these plastic molds can be reused numerous
times. Unlike the expensive setup and laborious processing required to
make the silicon wafers, this approach only requires a laser-jet
printer and a toaster oven, and can be completed within minutes.
Moreover, we can achieve multi-height designs within the device, which
typically requires a laborious and iterative process using standard
lithographic approaches.

Other novel methods have been developed as lower-cost alternatives to
photolithography, the gold standard for microfabrication and
microfluidic device creation. Duffy et al.2 first introduced rapid
prototyping of masters whereby they used printed transparencies to
replace the expensive chrome masks traditionally utilized in
photolithography. The authors demonstrated the advantages of using
rapid prototyping for masks over conventional photolithography and
micromachining. Despite its convenience, the method still requires the
use of expensive photoresist, high-resolution printing, and contact
lithography. Tan et al.3 obviated this issue by direct printing; they
photocopied designs onto transparencies to fabricate microfluidic
channel molds that ranged in height from 8 to 14 µm, depending on the
darkness setting of the photocopy machine. Liu et al.4 developed a
one-step direct-printing technique for the design and fabrication of
passive micromixers in microfluidic devices, with a maximum channel
height of 11 µm. Such shallow channels are adequate for many
microfluidic applications but not amenable for use with large
mammalian cells (>10 µm in diameter) as well as other applications,
such as flowing chemotactic gradients across adherent cells in a
channel with minimal shearing.5

While Lago et al.6 introduced a way to circumvent the height
limitation of single-layer ink by printing up to four times using a
thermal toner transfer method onto a glass substrate, the maximum
height obtained with this approach was 25 µm. Vullev et al.7
demonstrated a non-lithographic fabrication approach of microfluidic
devices by printing positive-relief masters with a laser-jet printer
for detecting bacterial spores; the height of the channels, which is
likewise dependent on the height of the ink, is limited to between 5
and 9 µm. To achieve deep channels, McDonald et al.8 introduced the
use of solid object printing (SOP) to make PDMS molds in
thermoplastics. However, despite their versatility, solid object
printers are considerably costly ($50000.)

Furthermore, the majority of these methods (as well as conventional
photolithography) produce rectangular cross section channels.
Pneumatic valves, first introduced by Quake and co-workers,9 important
for many microfluidic applications, require microfluidic channels to
be rounded such that they can be completely sealed upon valve closure.
Achieving rounded microfluidic channels using typical
photolithographic techniques, however, is complicated and requires an
extra re-flow step of the photoresist at high temperatures. Most
recently, Chao et al.10 demonstrated an elegant rapid prototyping
approach, coined microscale plasma templating (µPLAT), using water
molds. This technique enables the creation of rounded channels that
are difficult to make with photolithography, but still requires
micromachined masks and plasma activation.

In this Note, we present a simple method to fabricate microfluidic
channel molds that are inherently rounded. We demonstrate the ability
to create molds by printing at a larger scale and then shrinking down
more than 60% by leveraging the inherent property of thermoplastics.
..."


More information about the FoRK mailing list