Rapid fabrication of a poly(dimethylsiloxane) microfluidic

capillary gel electrophoresis system utilizing high precision

machining

Dong S. Zhao, Binayak Roy, Matthew T. McCormick, Werner G. Kuhr† and Sara A.
Brazill*

Department of Chemistry, University of California, Riverside, CA 92521, USA

Received 15th January 2003, Accepted 4th March 2003

First published as an Advance Article on the web 18th March 2003

In this work, we demonstrate a rapid protocol to address one of the major barriers that exists in the fabrication of
chip devices, creating the micron-sized structures in the substrate material. This approach makes it possible to
design, produce, and fabricate a microfluidic system with channel features > 10 µm in poly(dimethylsiloxane)
(PDMS) in under 8 hours utilizing instrumentation common to most machine shops. The procedure involves the
creation of a master template with negative features, using high precision machining. This master is then
employed to create an acrylic mold that is used in the final fabrication step to cast channel structures into the
PDMS substrate. The performance of the microfluidic system prepared using this fabrication procedure is
evaluated by constructing a miniaturized capillary gel electrophoresis (micro-CGE) system for the analysis of
DNA fragments. Agarose is utilized as the sieving medium in the micro-CGE device and is shown to give
reproducible (RSD (n = 34) ~ 5.0%) results for about 34 individual separations without replenishing the gel. To
demonstrate the functionality of the micro-CGE device, a DNA restriction ladder (spanning 26–700 base pairs)
and DNA fragments generated by PCR are separated and detected with laser-induced fluorescence (LIF). The
microchip is shown to achieve a separation efficiency of 2.53 3 105 plates m21.

Introduction relatively fast fabrication procedures, and unnecessary clean

The desire to analyze minute quantities of sample in a high
throughput and sensitive manner has motivated research into the
miniaturization of analytical instrumentation. The necessity for
high-throughput screening in areas such as: genetic analysis,
drug discovery, clinical diagnostics, and environmental mon-
itoring, has further intensified the interest in developing
effective miniaturized analytical devices. A miniaturized total
analysis system (mTAS) is a device fully integrated with all
functions necessary for a particular analysis.1–5 Recently,
miniaturized devices with integrated sample introduction,6
analyte pre-concentration,7,8 chemical reactions,9–12 separa-
tion,13–15 and detection16,17 have been investigated. The
successful development of a true mTAS promises ultra-high-
throughput screening capabilities along with a dramatic reduc-
tion in sample, reagent, and solvent requirements over conven-
tional bench-top analysis techniques.
In the past decade, the majority of microfluidic systems were
fabricated with glass or oxidized silicon (Si/SiO2) sub-
strates.18–22 The initial dominance of glass and silicon as
substrates for mTAS was partly due to pre-existing fabrication
techniques such as photolithography, dry and wet etching, and
laser cutting, which were developed and optimized by the
semiconductor industry. These techniques have been optimized
for the production of computer microchips and offer precision at
the micron to sub-micron scale.23 Although glass substrates are
well-suited for use in mTAS, drawbacks exist and are associated
with the laborious fabrication procedures and specialized
instrumentation/facilities necessary for production.24,25
The disadvantages associated with glass and silicon have
inspired research into utilizing polymers for the generation of
microfluidic devices.26–29 Polymer substrates are an attractive
alternative for mTAS fabrication due to the cost effectiveness,
† Current address: ZettaCore Inc., Denver, CO 80222, USA.
room environment. Generally, there are three fabrication
techniques used to create polymer microchips: casting,30,31
injection molding,32,33 and hot embossing.26,34–40 Poly(dime-
thylsiloxane) (PDMS) is a popular polymer due to its hydro-
phobic surface characteristics, excellent molding properties,
and facile sealing (either reversible or irreversible) to numerous
substrates.26,41 Additionally, PDMS possesses desirable optical
characteristics, a relatively low refractive index (n = 1.430),
and an absorbance cut-off above 230 nm.
PDMS can reproduce sub-micron features when molded or
cast onto a master template. The master contains positive relief
features that translate into the fluidic chambers and three-
dimensional structures when PDMS is molded onto it. Ideally,
the master molds would be produced in a rapid, cost effective
manner to allow efficient high throughput prototyping of the
mTAS. Several groups have worked on this particular goal in the
past decade. Effenhauser and co-workers first suggested using
an etched silicon wafer as a master to obtain narrow PDMS
channels, which they utilized to electrophoretically separate and
analyze DNA restriction fragments.34 Additionally, Whitesides
et al. have published extensive work on methods to mold
PDMS, for example they suggested a modified procedure of the
silicon master mold which is based on a “stamp” concept.42,43 In
this method, they converted the desired microfluidic design
from a CAD program onto a transparency utilizing a high-
resolution printer. The transparency was then used as a mask to
photolithographically generate a PDMS master with positive
photoresist.37
An alternative procedure for PDMS mask production is
demonstrated in this paper. The procedure is based on molding
PDMS against an acrylic master generated from a block of
aluminium cut with a high precision mill. The master fabrica-
tion procedure presented provides an easy and inexpensive
approach to make mTAS without the need to use photolitho-
graphic procedures. Micro-machining technology has provided
tools for fabricating miniaturized, three-dimensional structures

DOI: 10.1039/b300577a Lab Chip, 2003, 3, 93–99 93

This journal is © The Royal Society of Chemistry 2003
capable of performing clinically relevant analyses.44–48 Micro-
machining technology typically involves multiple steps such as:
microstructure design, master fabrication, chip fabrication,
assembly, and bonding. This technology offers an opportunity
to develop miniaturized instrumentation with a highly auto-
mated procedure and rapid turn-around time.48–51 The mold
fabrication procedure is validated through the construction of a
microchip capillary gel electrophoresis (micro-CGE) system for
the separation and analysis of DNA fragments. In our device,
agarose gel is used as the sieving matrix for DNA separation and
provides a robust separation where in excess of 30 runs can be
performed without regenerating the sieving medium.
Experimental
Chemicals and materials
pBluescript SK+ phagemid (pBSK+) was purchased from
Stratagene (La Jolla, CA). Taq DNA polymerase, standard 103
PCR buffer, TurboNAR1, PhiX 174 DNA–Hae III digest, and
T4 Polynucleotide Kinase were purchased from Promega
(Madison, WI). The restriction enzymes, Msp1 and NAR1, and
T4 DNA ligase were purchased from New England Biolabs
(Beverly, MA). dNTP and primer pairs labeled with 5-carboxy-
fluorescein (FAM) were supplied by Applied Biosystem (Foster
City, CA). The primer pairs used for the PCR amplification of
pBSK+ are as follows: 1) forward T3 primer: 5A-FAM-AAC
CAA TAG GCC GAA ATC GG (T3 FAM), 2) reverse 159 bp
primer: 5A-GTG CCG TAA AGC ACT AAA TC, 3) reverse 268
bp primer: 5A-TGG CAA GTG TAG CGG TCA CG, 4) reverse
402 bp primer: 5A-TTA CGC CAG CTG GCG AAA GG, 5)
reverse 524 bp primer: 5A-ACT CAC TAT AGG GCG AAT TG,
and 6) reverse 721 bp primer: 5A-AAC CAA TAG GCC GAA
ATC GG. PDMS elastomer was purchased from Dow Corning
Sylgard 184 (K. R. Anderson Co., Santa Clear, CA) as a two-
component kit: a base and a curing agent.
Generation of FAM labeled PCR fragments
Double stranded DNA fragments, 159, 268, 402, 524, and 721
bps, were generated from the pBSK+ template utilizing PCR.
The PCR reaction was carried out in a 50 µL reaction mixture
with the following final composition: 13 PCR buffer (50 mM
KCl and 10 mM Tris/HCl pH 8.3); 1.0 mM MgCl2; 200 µM
dNTPs; 330 ng of each primer; 2.5 units of Taq polymerase; 50
ng pBSK+ template; and sterile water. The PCR amplification
reaction was performed in a GeneAmp PCR system (Perkin-
Elmer model 2400) under the following conditions: denatura-
tion at 94 °C for 4 min; followed by 45 cycles of: 94 °C for 30
s, 55 °C for 30 s, 72 °C for 45 s, concluding with a 7 min 72 °C
extension step. After PCR was complete, the generation of the
correct product was confirmed through slab gel electrophoresis.
The PCR fragments and PhiX 174 DNA marker ladder were run
on a 2% agarose gel, with ethidium bromide staining, to ensure
the correct length fragments were generated.
PCR fragment clean-up
The PCR fragments were purified utilizing the MinElute Gel
Extraction kit (Qiagen, Valencia, CA). The DNA fragments
were excised from the agarose gel and placed into a 2 ml vial.
94 Lab Chip, 2003, 3, 93–99
The instructions on the kit were followed to bind, wash, and
elute the PCR fragments.
Generation of a FAM labeled DNA ladder
A DNA ladder, with a 2-basepair 5A overhang, was generated
from the pBSK+ plasmid utilizing the Msp1 restriction enzyme
to generate fragments that span from 26 bps to 700 bps. The
fragments generated were then ligated to the FAM T3 primer.
The simple and one step procedure involves a cocktail of
enzymes which was incubated with the plasmid to generate the
labeled fragments. Briefly, the reaction was carried out in a 100
µL reaction mixture including: 15 µg pBSK+, 10 µL Turbo-
NAR1 buffer, 3 µg T3 FAM (annealed double strand), 10 µl
ATP (10 mM), 2.5 µl NAR1 (10 units), 10 µl TurboNar1 (100
units), 2.5 µl MSP1 (50 units), 1.5 µl T4 Polynucleotide Kinase,
0.5 µl T4 DNA ligase, and sterile water. The reaction cocktail
was incubated at 37 °C overnight.
Fabrication of the aluminium master mold
The original design of the master mold was drawn on a CAD-
CAM program and then converted into a CNC-Mill program.
The substrate used for the mask was 6061 aluminium-alloy
because of its strength, softness, and ease of polishing.52 First,
the aluminium was polished down to 0.15 µm roughness, then
a CNC high precision mill-machine was used to cut the negative
channel features (Fig. 1A). Next, a piece of acrylic was clasped
onto the top of the aluminium mask and placed into an oven at
205 °C for 2 hours. This step resulted in the acrylic sinking
down into the channels made in the aluminium mask and
generated a positive feature in the acrylic surface (Fig. 1B).
After cooling, the acrylic wafer was gently released and ready
for PDMS casting.
PDMS chip fabrication
The layout of the microchip device used in this work is shown
in Fig. 1. The PDMS was polymerized by mixing 10+1 (w/w)
ratio Sylgard 184 with curing agent. The solution was stirred
and degassed under vacuum for 20 minutes. Then the viscous
solution was smoothly poured into the acrylic wafer. The wafer
was placed in an oven at 65 °C for 1 hour. Once cured, the
PDMS replica was easily peeled from the acrylic mold and
holes serving as reservoirs and providing access to the channels
were punched through the bulk material. To complete the
construction of the device, a PDMS/glass hybrid microchip was
constructed by covering the PDMS microchip with a glass plate
(4 3 3 inches). After placing the chip on a glass slide,
hermetically sealed micro-channels were readily formed by
mere adhesion without applying external force.
Polymer matrix filling
Two percent (w/v) agarose gel power (Sigma, St. Louis, MO)
was add into 1X TBE buffer and heated until a homogeneous
liquid was formed. The PDMS/glass channel surface was rinsed
with acetone and air-dried, then a drop of ethanol was added to
the top buffer reservoir (reservoir 2, Fig. 1A). Due to the
hydrophobic nature of the PDMS surface the ethanol solution
wicked into the channel structure without any applied pressure.
Once each channel was filled with ethanol, agarose gel solution,
maintained above its gelling temperature (10–15 °C), was added
to the waste reservoir (reservoir 4, Fig. 1A) and introduced into
the separation channel by applying pressure. The agarose gel
was pushed into the separation channel just until the gel front
reached the T-junction, then all of the reservoirs were filled with
running buffer (1X TBE) and the chip cooled to room
temperature. The agarose gel was matured at room temperature
and the chip was ready to use.
Instrumentation
Microchip-CGE separation was performed utilizing a computer
controlled high voltage power supply, built in-house and
modeled after James Landers group at the University of
Virginia. The power supply provided the necessary voltages for
sample injection and separation. The principles of sample
injection and device operation have been described in great
detail in the literature53–56 and will not be thoroughly discussed.
In this work, approximately 5–10 µL of sample was pipetted
into the sample reservoir (reservoir 1, Fig. 1A). By applying an
electrical field, from 200–400 V cm21, between reservoirs 1
and 3 the sample migrated through the sample channel and thus
filled the T area of the microchip. Following a predetermined
injection time, electrophoretic separation began by applying a
potential between reservoirs 2 and 4. The voltages and injection
time were controlled with a LabVIEW 5.1 program (National
Instruments, Austin, TX) operated on an IBM-compatible PC.
The DNA fragments, which were labeled with a fluorophore,
were detected utilizing a laser-induced fluorescence (LIF) set-
up. An Olympus microscope (Model: Vanox, Olympus Optical
Co. Ltd., Tokyo, Japan) equipped with a fluorescence phase
contrast attachment was utilized and the 488 nm line of an argon
ion laser (Uniphase, San Jose, CA) was used as the excitation
light source. Excitation light was reflected by a dichroic mirror
(DM 500) and focused onto the microchannel by means of a
microscope objective (Olympus, ULWD CDPlan 20PL 20 3
0.40). Fluorescence was collected back through the dichroic
mirror and detected with a photomultiplier tube (PMT)
(Hamamatsu, Bridgewater, NJ). The PMT signal was amplified
utilizing a Keithley 428 current amplifier (Keithley Instruments
Inc., Cleveland, OH), which converts the photo current into a
voltage with a gain of 106 V A21 and a 100 µs rise time. The
analog voltage output of the current amplifier was collected at
a rate of 10 Hz with a data acquisition board (PCI-4652E,
National Instruments, Austin, TX).
Fig. 1 Scheme illustrating the procedure for utilizing micro-machining to create a PDMS based device. (A) An image of a machined aluminium wafer
containing the negative channel relief features created in a CAD program; (B) an image of the acrylic mold created from the aluminium wafer; (C) an image
of the T-intersection of the sample and separation channel; (D) an image of the PDMS channel geometry.
Fluorescence imaging
Fluorescent imaging was performed with an epi-fluorescence
microscope (Zeiss Axioskop, Thornwood, NY) equipped with a
100 W Hg arc lamp for epi-illumination. Fluorescent images
were obtained by passing the light from the Hg arc lamp through
an excitation filter specific for the fluorescein absorption band
(494 nm) and fluorescence collected at wavelengths above 640
nm. All fluorescence images were acquired near the central
zone of the CCD with a camera gain setting of 1. All images
were collected and saved on a personal computer (PC) in a
darkened room with a cooled Princeton Instruments MicroMax
800-PB back-illuminated CCD system (800 3 1000 pixel, 16
bit per pixel) and ST-133 controller using Winview software.
Atomic force microscopy (AFM)
The surface roughness of the molded PDMS substrate was
determined utilizing contact-mode AFM performed on a
Burleigh Personal AFM, ARIS 3500. Silicon tips (100 nm,
pyramidal shape), also supplied from Burleigh (Burleigh Ins.
Co., Fishers, NY), were used as cantilevers. The reference force
of the cantilever, all gains, and filters were optimized prior to
each scan. Post-processing of the images was done using True
Image SPM software (Burleigh Instrument Co., Fishers, NY).
Results and discussion
Characteristics of the PDMS molding technique
Modern micromachining techniques such as: sawing, cutting,
and milling, are capable of producing structures in the range of
a few µm and larger in a variety of materials, particularly
stainless steel and aluminium. Recently, Madou et al. demon-
strated the utility of utilizing CNC-machining to rapidly create
prototypes on a plastic polycarbonate CD platform.57–59 The
microfluidic pattern was either; 1. machined directly into the
polycarbonate substrate or 2. a master mold with positive
features was created out of aluminium and PDMS was used to
create the microfluidic substrate. It was found that the
machining technique produced large relatively rough features
Lab Chip, 2003, 3, 93–99 95
with a low aspect ratio (depth/width), which were not optimum
for device performance. However, they did state that the method
enabled them to conduct rapid experiments to assess the
characteristics of different prototype designs.58
The machining process in this work involves milling a
negative feature into the aluminium substrate (shown in Fig.
1A), which is easier and can produce smaller features compared
to machining aluminium into a positive mold. The size limit for
the tool used to mill the aluminium substrate is approximately
10 µm, but this dimension can be scaled down further utilizing
more sophisticated machinery, e.g. electronic discharge ma-
chine (EDM). One nice advantage of the machining technique is
that there is no limitation on substrate size or channel length that
can be generated. Theoretically any size chip can be produced.
Using this procedure, several different acrylic wafers have been
fabricated with different patterns, such as a T-shape (Fig. 1) and
Y-shape (not shown), within 8 hours. Thus, with this technique
it is relatively easy to prototype many different designs to
determine which provides the optimum characteristics for a
particular application.
Once the first negative master is created in aluminium, the
positive relief mold is constructed from acrylic (shown in Fig.
1B). Acrylic is a heat insulator, which has a glass transition
temperature (Tg) of 105 °C and a crystalline melting tem-
perature (Tm) of 200 °C.60 Therefore, when acrylic is heated
above its Tm only the acrylic surface close to the aluminium
mask can reach the set temperature. As a result, the surface of
the acrylic is gradually transformed into a viscous liquid and
sinks into the negative features on the aluminium mask. The
depth of the channel created in the acrylic is strongly dependent
on the heating time. We found that an acrylic mask heated above
the Tm for approximately 2 hours results in channel heights from
7 to 20 µm. The work presented herein utilizes an acrylic mask
with a defined depth of 17 µm.
The PDMS chip is obtained in the standard fashion of casting
the polymer against a mold to yield elastomeric replicas
containing networks of channels. The PDMS channel profile at
the T-intersection is depicted in Fig. 1C. As can be seen from
the image, the intersection of the separation and injection
channel reveals straight edges with 90° angles formed. The
geometric volume of the T-channel intersection shown in Fig.
1C is calculated to be 44 pL. Fig. 1D is an image of the PDMS
substrate cut through the channel feature, which provides a side
view of the channel. The channel created in the PDMS replica
has a trapezoidal geometry, with a width of about 64 µm on top
and 38 µm on the bottom, and 17 µm in depth.
It is important to ensure the acrylic master generates
reproducible features through out the entire PDMS substrate.
Table 1 summaries the channel dimensions for ten different
PDMS chips cast from the same acrylic mold. The measure-
ments are generated by averaging the dimensions at five
different locations throughout each chip. Some of the channel
Fig. 2 Characterization of the molded PDMS chip surface. (A) A surface plot of the fluorescence intensity of a channel filled with 100 nM fluorescein; (B)
an AFM image of the PDMS chip surface. The relative roughness is locally about 0.1 µm and about 0.5 µm from peak-to-peak across the whole surface.
96 Lab Chip, 2003, 3, 93–99
walls are found to have defects, which could be a result of
improperly releasing the PDMS from the acrylic wafer.
However, it can be seen from Table 1 that the width of unsealed
channels correlated quite well with the acrylic master, with a
RSD of less than 3% for both the same chip and between chips.
Additionally, the depth of the channels has a relatively low
variability, RSD of less than 6%. It appears that the acrylic
master can be used for reproducible PDMS chip fabrication.
In addition to channel dimensions, surface roughness is
another important property in microfluidic systems. In order to
examine the roughness inside the PDMS channel, the channel is
filled with a solution of 100 nM fluorescein and the fluores-
cence intensity inside the channel is imaged. Fig. 2A shows the
surface profile of the fluorescence intensity versus length of the
channel. The surface plot indicates that the surface of the PDMS
channel is fairly uniform. The channel geometry illustrated in
the fluorescence image does not exactly correspond to the
trapezoidal geometry seen in the visible image of Fig. 1D. We
believe the reason for this deviation has to do with imprecise
microscope alignment and focusing when the fluorescence
image was acquired.
The PDMS surface roughness, surrounding the channel, is
measured with an AFM. The AFM image (Fig. 2B) indicates
that the PDMS surface is relatively rough. This roughness
reflects the defects present on the acrylic master surface because
PDMS accurately reproduces the surface of the mold used. The
relative roughness is locally about 0.1 µm and about 0.5 µm
from peak-to-peak across the whole surface. These defects are
due to the degradation of the acrylic caused by the heat
variability of the oven. In our study, we found that carefully
controlling the oven temperature will significantly reduce the
roughness up to three-fold. Additionally, ensuring the alumin-
ium surface of the negative mold is polished before the acrylic
mold is cast can reduce the roughness. In our studies, we did not
find the roughness surrounding the separation channel to cause
any adverse effects on sealing or the separation efficiency.
Table 1 Dimensions of created channels (n = 10)
Chip no. Width/µm Depth/µm
1 63.08 ± 1.44 16.89 ± 0.50
2 63.08 ± 1.29 17.71 ± 0.72
3 64.45 ± 1.87 17.08 ± 0.55
4 63.81 ± 0.37 17.57 ± 0.74
5 65.82 ± 0.76 16.84 ± 0.58
6 65.27 ± 1.11 17.37 ± 0.68
7 65.94 ± 1.11 17.76 ± 0.79
8 64.24 ± 1.64 17.62 ± 0.90
9 64.24 ± 1.50 17.71 ± 0.99
10 64.61 ± 1.85 17.23 ± 0.63
Chip-to-chip 64.46 ± 0.43 17.38 ± 0.23
PCR fragment separation on microchip-CGE
Agarose gel is a commercially available product that can be
used for efficient DNA separation and sizing applications. The
PDMS channel is first pre-wetted with ethanol and then easily
filled with a 2% agarose solution, which is held above its gelling
temperature. The agarose solution is pushed into the channel
until it reaches the T-intersection and then allowed to cool to
room temperature where it solidifies into a gel. We believe the
viscosity of the cooled agarose helps to reduce the diffusional
broadening of the sample plug during injections. Shown in Fig.
3 is the electropherogram of a mixture of 5 PCR fragments
labeled with the FAM fluorophore. The sample is separated in
1X TBE buffer and all 5 DNA PCR fragments are separated
with baseline resolution in under 200 seconds.
Table 2 summarizes the results from 6 separations of the PCR
fragments performed on the same chip without refilling the
agarose gel. A blank run is required between separations to
eliminate residual sample from previous runs. The RSD (n = 6)
of the migration times for each fragment is under 1.4%, which
indicates that the agarose gel maintains its sieving capability
throughout multiple injections. In addition to the highly
reproducible migration times, the separation efficiency for the
microchip is at least 205 000 theoretical plates m21. This
separation efficiency is comparable to or better than most
published micro-CGE separations results.37,61–66 The ability to
perform reproducible and efficient separation of DNA frag-
ments without the need to replenish the sieving medium will not
only increase throughput, but will also reduce reagent consump-
tion and cost.
Multiple separations of a DNA ligation ladder on the
microchip-CGE system
The agarose filled microchip-CGE was also tested with a
fluorophore labeled ladder. The protocol used in this work is
designed to join almost any 5A labeled oligomer to DNA
Fig. 3 Electropherogram of the PCR fragments labeled with FAM on the
microchip. Matrix: 2% agarose (type VIII), running buffer: 1X TBE; E =
110 V cm21.
Table 2 Separation efficiency and migration time reproducibility for the
four PCR products
Fragment size/bp RSD (%) (n = 6) Plate number/plates m21
159 1.33 248 000
268 1.31 205 000
402 1.44 253 000
524 1.30 271 000
721 1.37 290 000
fragments generated with the Msp1 restriction enzyme. The
procedure is simple and one step, in that a cocktail of enzymes
is incubated with a plasmid to generate labeled fragments. In
this procedure, pBSK+ plasmid is digested by the Msp1 enzyme
present in the cocktail. Msp1 digest of pBSK+ generates a DNA
ladder that spans, in a fairly uniform manner, the range of sizes
expected from a sequencing reaction (from 26 bps to 700 bps).
In addition, two of the fragments are only 10 bases apart and can
serve as an internal control to help assign molecular weights to
the other fragments. Msp1 generates DNA fragments with a 5A
overhang that allows attachment to almost “any” oligomer
sequence that is labeled at the 5A end.
Shown in Fig. 4 is the separation of the pBSK+ ladder labeled
with a FAM fluorophore performed on the PDMS micro-CGE.
To further illustrate the stability of the agarose sieving matrix,
34 separations of the ladder are performed on the same
microchip. Fig. 4A is the electropherogram from the first run
and Fig. 4B is electropherogram number 34. The run-to-run
variations were small, with the largest RSD in migration time
being less than 5% (n = 34). It can be seen from the 34th
separation that the signal to noise value is much smaller
compared to the initial separation. We are not sure if this is due
to the detection system or if the analyte concentration is
dramatically lowered due to the number of runs. A detailed
study into the characteristics of multiple separations performed
with the agarose sieving matrix is currently underway in our
laboratory. Regardless, the ability to perform 34 sequential
separations without dramatic loss in reproducibility or separa-
tion efficiency will provide improvements in throughput,
sample/reagent consumption, and contamination reduction.
Conclusions
We have developed a system for fabricating networks of
channels ( > 10 µm width), which is based on the prototyping of
masters using high-precision machining and molding of PDMS.
Fig. 4 Electropherogram of the FAM labeled Msp1 digest of pBSK+
plasmid on the microchip. Matrix: 2% agarose gel (type VIII), running
buffer: 1X TBE; E = 100 V cm21. (A) run 1, (B) run 34 without
regeneration of the agarose.
Lab Chip, 2003, 3, 93–99 97
The entire fabrication process from concept to realization takes
less than 8 h and thus can be used to rapidly prototype
microfluidic designs. The protocol to create an acrylic wafer
presented herein possesses several advantages including: (1) the
reduction in production time and cost compared to silicon or
glass wafers; (2) the utilization of fairly simple instrumentation
and facilities. The only necessary tool is a high precision miller,
which is present in most machine shops; (3) there is no
limitation on the size of the created device.
With the miniaturized CGE system both PCR generated
fragments and a DNA ladder generated from a ligation reaction
were successfully separated. Additionally, the agarose sieving
matrix could be utilizing for multiple runs (up to 34) without
refilling the channel. The reproducibility in migration time with
the same chip through multiple injections was on the order of
5% and the efficiency of the separations was not compromised
throughout the 34 runs. Future work will focus on fully
characterizing the durability of the agarose matrix throughout
multiple runs.
Acknowledgement
This work was supported by the National Institutes of Health
(1R21HG01828–01). The authors would like to thank James P.
Landers and Zhili Huang of the University of Virginia for
providing the 4-channel high voltage power supply circuit
schematic.
References
1 E. Dempsey, D. Diamond, M. R. Smyth, G. Urban, G. Jobst, I. Moser,
E. M. J. Verpoorte, A. Manz, H. M. Widmer, K. Rabenstein and R.
Freaney, Design and development of a miniaturized total chemical
analysis system for on-line lactate and glucose monitoring in
biological samples, Anal. Chim. Acta, 1997, 346, 341–349.
2 Y. Xu, J. C. T. Eijkel and A. Manz, Advances of miniaturized total
chemical analysis system and its techniques, Chem. J. Chin. Univ.,
2000, 21, 1028–1034.
3 S. C. Jakeway, A. J. de Mello and E. L. Russell, Miniaturized total
analysis system for biological analysis, Fresenius J. Anal. Chem.,
2000, 366, 525–539.
4 L. Bousse, C. Cohen, T. Nikiforov, A. Chow, A. R. Kopf-Sill, R.
Dubrow and J. W. Parce, Electrokinetically controlled microfluidic
analysis systems, Annu. Rev. Biophys. Biomol. Struct., 2000, 29,
155–181.
5 R. C. McGlennen, Miniaturization technologies for molecular
diagnostics, Clin. Chem., 2001, 47, 393–402.
6 Q. Fang, G.-M. Xu and Z.-L. Fang, A High-Throughput Continuous
Sample Introduction Interface for Microfluidic Chip-based Capillary
Electrophoresis Systems, Anal. Chem., 2002, 74, 1223–1231.
7 J. Khandurina, S. C. Jacobson, L. C. Waters, R. S. Foote and J. M.
Ramsey, Microfabricated Porous Membrane Structure for Sample
Concentration and Electrophoretic Analysis, Anal. Chem., 1999, 71,
1815–1819.
8 Y. C. Lin, H. C. Ho, C. K. Tseng and S. Q. Hou, A poly-
methylmethacrylate electrophoresis microchip with sample pre-
concentrator, J. Micromech. Microeng., 2001, 11, 189–194.
9 G. H. Seong, W. Zhan and R. M. Crooks, Fabrication of
microchambers defined by photopolymerized hydrogels and weirs
within microfluidic systems: Application to DNA hybridization, Anal.
Chem., 2002, 74, 3372–3377.
10 S. V. Tillib, B. N. Strizhkov and A. D. Mirzabekov, Integration of
Multiple PCR Amplifications and DNA Mutation Analyses by Using
Oligonucleotide Microchip, Anal. Biochem., 2001, 292, 155–160.
11 M. N. Slyadnev, Y. Tanaka, M. Tokeshi and T. Kitamori, Photo-
thermal temperature control of a chemical reaction on a microchip
using an infrared diode laser, Anal. Chem., 2001, 73, 4037–4044.
12 Y. Tanaka, M. N. Slyadnev, A. Hibara, M. Tokeshi and T. Kitamori,
Non-contact photothermal control of enzyme reactions on a micro-
chip by using a compact diode laser, J. Chromatogr. A, 2000, 894,
45–51.
98 Lab Chip, 2003, 3, 93–99
13 A. Manz, N. Graber and M. Widmer, Miniaturized total chemical
analysis system – A novel concept for chemical sensing, Sens.
Actuators B, 1990, 1, 244–248.
14 J. W. Hong, K. Hosokawa, T. Fujii, M. Seki and I. Endo,
Microfabricated polymer chip for capillary gel electrophoresis,
Biotechnol. Prog., 2001, 17, 958–962.
15 W. Xu, K. Uchiyama, T. Shimosaka and T. Hobo, Fabrication of
polyester microchannels and their applications to capillary electro-
phoresis, J. Chromatogr. A, 2000, 907, 279–289.
16 S. Qi, X. Liu, S. Ford, J. Barrows, G. Thomas, K. Kelly, A.
McCandless, K. Lian, J. Goettert and S. A. Soper, Microfluidic
devices fabricated in poly(methyl methacrylate) using hot-embossing
with integrated sampling capillary and fiber optics for fluorescence
detection, Lab Chip, 2002, 2, 88–95.
17 S. C. Wang and M. D. Morris, Plastic microchip electrophoresis with
analyte velocity modulation. Application to fluorescence background
rejection, Anal. Chem., 2000, 72, 1448–1452.
18 P. Wilding, J. Pfahler, H. H. Bau, J. N. Zemel and L. Kricka,
Manipulation and flow of biological fluids in straight channels
micromachined in silicon, J. Clin. Chem., 1994, 40, 43–47.
19 R. Furlan and J. N. Zemel, Behavior of microfluidic amplifiers, Sens.
Actuators A, 1996, 51, 239–246.
20 R. M. Guijt, E. Baltussen, G. Van der Steen, R. B. M. Schasfoort, S.
Schlautmann, H. A. H. Billiet, J. Frank, G. W. K. Van Dedem and A.
Van den Berg, New approaches for fabrication of microfluidic
capillary electrophoresis devices with on-chip conductivity detection,
Electrophoresis, 2001, 22, 235–241.
21 J. Khandurina, T. E. McKnight, S. C. Jacobson, L. C. Waters, R. S.
Foote and J. M. Ramsey, Integrated system for rapid PCR-based
DNA analysis in microfluidic devices, Anal. Chem., 2000, 72,
2995–3000.
22 Y. Murakami, T. Takeuchi, K. Yokoyama, E. Tamiya, I. Karube and
M. Suda, Intergration of enzyme-immobilized column with electro-
chemical flow cell using micromachining techniques for a glucose
detection system, Anal. Chem., 1993, 65, 2731–2735.
23 H. Yoon, G. W. Cha, C. Yoo, N. J. Kim, K. Y. Kim, C. H. Lee, K. N.
Lim, K. Lee, J. Y. Jeon, T. S. Jung, H. Jeong, T. Y. Chung, J. Kim and
S. I. Cho, A 2.5-V, 333-Mb/s/pin, 1-Gbit, double-data-rate synchro-
nous DRAM, IEEE J. Solid-State Circuits, 1999, 34, 1589–1599.
24 A. J. Nijdam, J. W. Berenschot, J. van Suchtelen, J. G. E. Gardeniers
and M. Elwenspoek, Velocity sources as an explanation for
experimentally observed variations in Si{111} etch rates, J. Micro-
mech. Microeng., 1999, 9, 135–138.
25 B. He, L. Tan and F. Regnier, Microfabricated Filters for Micro-
fluidic Analytical Systems, Anal Chem., 1999, 71, 1464–1468.
26 H. Beck and C. Gartner, Polymer microfabrication methods for
microfluidic analytical applications, Electrophoresis, 2000, 21,
12–26.
27 S. L. R. Barker, M. J. Tarlov, H. Canavan, J. J. Hickman and L. E.
Locascio, Plastic Microfluidic Devices Modified with Polyelectrolyte
Multilayers, Anal. Chem., 2000, 72, 4899–4903.
28 J. Monahan, A. A. Gewirth and R. G. Nuzzo, A Method for Filling
Complex Polymeric Microfluidic Devices and Arrays, Anal. Chem.,
2001, 73, 3193–3197.
29 J. C. McDonald and G. M. Whitesides, Poly(dimethylsiloxane) as a
Material for Fabricating Microfluidic Devices, Acc. Chem. Res.,
2002, 35, 491–499.
30 T. Fujii, PDMS-based microfluidic devices for biomedical applica-
tions, Microelectron. Eng., 2002, 61/62, 907–914.
31 T. McCreedy, Rapid prototyping of glass and PDMS microstructures
for micro total analytical systems and micro chemical reactors by
microfabrication in the general laboratory, Anal. Chim. Acta, 2001,
427, 39–43.
32 T. Boone, Z. H. Fan, H. Hooper, A. Rico, H. D. Tan and S. Williams,
Plastic advances microfluidic devices, Anal Chem., 2002, 74, 78A–
86A.
33 A. P. Sassi, I. D. Cruzado, T. Björnson, H. H. Hooper and A. P.
Paulus, Rapid, parallel separations of D1S80 alleles in a plastic
microchannel chip, J. Chromatogr. A, 2000, 894, 203–217.
34 C. S. Effenhauser, G. J. M. Bruin, A. Paulus and M. Ehrat, Integrated
Capillary Electrophoresis on Flexible Silicone Microdevices: Analy-
sis of DNA Restriction Fragments and Detection of Single DNA
Molecules on Microchips, Anal. Chem., 1997, 69, 3451–3457.
35 L. Martynova, L. E. Locascio, M. Gaitan, G. W. Kramer, R. G.
Christensen and W. A. MacCrehan, Fabrication of Plastic Micro-
fluidic Channels by Imprinting Methods, Anal. Chem., 1997, 69,
4783–4789.
36 R. M. McCormick, R. J. Nelson, M. G. Alonso-Amigo, D. J.
Benvegnu and H. H. Hooper, Microchannel Electrophoretic Separa-
tions of DNA in Injection-Molded Plastic Substrates, Anal. Chem.,
1997, 69, 2626–2630.
37 D. C. Duffy, J. C. McDonald, J. A. Schueller and G. M. Whitesides,
Rapid Prototyping of Microfluidic Systems in Poly(dimethylsilox-
ane), Anal. Chem., 1998, 70, 4974–4984.
38 Z. Meng, S. Qi, S. A. Soper and P. A. Limbach, Interfacing a
polymer-based micromachined device to a nanoelectrospray ioniza-
tion Fourier Transform ion cyclotron resonance mass spectrometer,
Anal Chem., 2001, 73, 1286–1291.
39 A. Gerlach, G. Knebel, A. E. Guber, M. Heckele, D. Herrmann, A.
Muslija and T. Schaller, Microfabrication of single-use plastic
microfluidic devices for high-throughput screening and DNA analy-
sis, Microsystem Technol., 2002, 7, 265–268.
40 S. Qi, X. Liu, S. Ford, J. Barrows, G. Thomas, K. Kelly, A.
McCandless, K. Lian, J. Goettert and S. A. Soper, Microfluidic
devices fabricated in poly(methyl methacrylate) using hot-embossing
with integrated sampling capillary and fiber optics for fluorescence
detection, Lab Chip, 2002, 2, 88–95.
41 J. C. McDonald, D. C. Duffy, J. R. Anderson, D. T. Chiu, H. Wu, O.
J. A. Schueller and G. M. Whitesides, Fabrication of microfluidic
systems in poly(dimethylsiloxane), Electrophoresis, 2000, 21,
27–40.
42 E. Kim, Y. Xia and G. M. Whitesides, Polymer microstructures
formed by molding in capillaries, Nature, 1995, 376, 581–584.
43 R. J. Jackman, J. L. Wilber and G. M. Whitesides, Fabrication of
submicrometer features on curved substrates by microcontact
printing, Science, 1995, 269, 664–666.
44 S. Shoji, M. Esashi and T. Matsuo, Prototype miniature blood gas
analyzer fabricated on a silicon wafer, Sens. Actuators, 1988, 14,
101–107.
45 A. Manz, J. C. Fettinger, E. Verpoorte, H. Lüdi, H. M. Widmer and
D. J. Harrison, Micromachining of monocrystalline silicon and glass
for chemical analysis systems: a look into next century’s technology
or just a fashionable craze?, Trends Anal. Chem., 1991, 10,
144–149.
46 D. J. Harrison, K. Fluri, K. Seiler, Z. Fan, C. S. Effenhauser and A.
Manz, Micromachining a miniaturized capillary electrophoresis-
based chemical analysis system on a chip, Science, 1993, 261,
895–897.
47 N. Chiem and D. J. Harrison, Microchip-based capillary electro-
phoresis for immunoassays: analysis of monoclonal antibodies and
theophylline, Anal. Chem., 1997, 69, 373–378.
48 J. Cheng, P. Fortina, S. Surret, L. J. Kricka and P. Wilding,
Microchip-based devices for molecular diagnosis of genetic diseases,
Mol. Diagn., 1996, 1, 183–200.
49 D. J. Harrison, A. Manz, Z. Fan, H. Lüdi and H. M. Widmer,
Capillary electrophoresis and sample injection systems integrated on
a planar glass chip, Anal. Chem., 1992, 64, 1926–1932.
50 M. T. McCaman, P. Murakami, E. Pungor, K. M. Hahnenberger and
W. S. Hancock, Analysis of recombinant adenoviruses using an
integrated microfluidic chip-based system, Anal. Biochem., 2001,
291, 262–268.
51 L. B. Koutny, D. Schmaizing, T. A. Taylor and M. Fuchs, Microchip
electrophoretic immunoassay for serum cortisol, Anal. Chem., 1996,
68, 18–22.
52 http://www.metalsonline.com/propertypages/6061.asp.
53 J. P. Alarie, S. C. Jacobson and J. M. Ramsey, Electrophoretic
injection bias in a microchip valving scheme, Electrophoresis, 2001,
22, 312–317.
54 C. X. Zhang and A. Manz, Narrow sample channel injectors for
capillary electrophoresis on microchips, Anal. Chem., 2001, 735,
2656–2662.
55 L.-M. Fu, R.-J. Yang, G.-B. Lee and H.-H. Liu, Electrokinetic
Injection Techniques in Microfluidic Chips, Anal. Chem., 2002, 74,
5084–5091.
56 G. Ocvirk, M. Munroe, T. Tang, R. Oleschuk, K. Westra and D. J.
Harrison, Electrokinetic control of fluid flow in native poly-
(dimethylsiloxane) capillary electrophoresis devices, Electrophor-
esis, 2000, 21, 107–115.
57 M. J. Madou, Y. Lu, S. Lai, C. G. Koh, L. J. Lee and B. R. Werner,
A design on a CD disc for 2-point calibration measurement, Sens.
Actuators A, 2001, 91, 301–306.
58 M. J. Madou, L. J. Lee, K. W. Koelling, S. Dauner, S. Lai, C. G. Koh,
Y.-J. Juang, L. Yu and Y. Lu, Design and fabrication of polymer
microfluidic platforms for biomedical applications, ANTEC-SPE
59th, 2001, 2534–2538.
59 M. J. Madou, L. J. Lee, S. Dauner, S. Lai and C.-H. Shih, Design and
fabrication of CD-like microfluidic platforms for diagnostics:
Microfluidic functions, Biomed. Microdev., 2001, 3, 245–254.
60 G. Odian, Principles of Polymerization, John Wiley & Sons, New
York, 1991, p. 32.
61 R. S. Martin, A. J. Gawron and S. Lunte, Dual-electrode electro-
chemical detection for poly(dimethylsiloxane)-fabricated capillary
electrophoresis microchips, Anal. Chem., 2000, 72, 3196–3202.
62 W. Xu, K. Uchiyama, T. Shimosaka and T. Hobo, The application of
polyester microchip to capillary electrophoresis, Chem. Lett., 2000,
762–763.
63 I. Rodriguez, L. J. Jin and S. F. Y. Li, High-speed chiral separations
on microchip electrophoresis devices, Electrophoresis, 2000, 21,
211–219.
64 J. Khandurina, T. E. McKnight, S. C. Jacobson, L. C. Waters, R. S.
Foote and J. M. Ramsey, Integrated System for Rapid PCR-based
DNA analysis in Microfluidic Devices, Anal. Chem., 2000, 72,
2995–3000.
65 R. S. Martin, A. J. Gawron and S. M. Lunte, Dual-Electrode
Electrochemical Detection for Poly(dimethylsiloxane)-fabricated
Capillary Electrophoresis Microchips, Anal. Chem., 2000, 72,
3196–3202.
66 C. S. Effenhauser, G. J. M. Bruin, A. Paulus and M. Ehrat, Integrated
Capillary Electrophoresis on Flexible Silicone Microdevices: Analy-
sis of DNA Restriction Fragments and Detection of Single DNA
Molecules on Microchips, Anal. Chem., 1997, 69, 3451–3457.
Lab Chip, 2003, 3, 93–99 99