LIQUID REAGENT STORAGE AND RELEASE FOR CENTRIFUGALLY OPERATED
LAB-ON-A-CHIP SYSTEMS BASED ON A BURSTABLE SEAL
J. Hoffmann1, D. Mark2, R. Zengerle1,2 and F. von Stetten1,2*
1Laboratory for MEMS Applications, Department of Microsystems Engineering (IMTEK),
University of Freiburg, Georges-Koehler-Allee 106, D-79110 Freiburg, Germany
2HSG-IMIT, Wilhelm-Schickard-Straße 10, D-78052 Villingen-Schwenningen, Germany
We present a new approach for the pre-storage and
release of liquid reagents in centrifugally operated Lab-on-
a-Chip (LoaC) cartridges. Liquids are stored in sealed
cavities which are separated from the fluidic system by a
weak-bonded interface. During centrifugal rotation, the
liquid exerts an inertial force onto the predetermined area.
This delaminates the sealing foil locally, resulting in a
fluidic connection to the downstream channel. Time-to-
release could be adjusted to a range between 31 s and
143 s by geometrical variations of the structure, enabling
time controlled release. In sum, the burstable seal is a
universal and robust valve: it is vapor tight, independent
on wetting properties of liquids, and time controllable.
Hence it excels siphon based valving concepts often used
in centrifugal microfluidics and is most valuable for the
design of LoaC cartridges for point of care applications.
Reagent storage, centrifugal microfluidics, Lab-on-a-Chip,
MEMS based microfluidic platforms represent a
promising technology for rapid in-vitro detection of
various microbiological and biochemical compounds .
Those systems can precisely handle minute amounts of
liquids enabling an automated performance of sensitive,
selective, and multiplexed bio-chemical assays on
miniaturized Lab-on-a-Chip (LoaC) cartridges.
The LoaC technology concatenates and miniaturizes
typical fluidic unit operations such as fluid transport,
metering, aliquoting, and mixing, providing the means to
shrink down complex laboratory functionality to chip-
sized substrates . Beside clinical applications, pocket-
size LoaC systems offer additional perspectives in
environmental and security testing, because they can be
battery operated, and applied by non-experts.
Despite significant advantages of LoaC systems, the
market penetration is still very limited, and, for example,
integration of reliable liquid reagent storage systems into
analytical cartridges is still a challenging issue .
Although storage of lyophilized reagents is frequently
reported, e.g. [4,5], concepts for pre-storing liquid reagents
in microfluidic cartridges only exist sporadically, e.g. [6-
8]. Nevertheless, the on-chip pre-storage of all required
reagents is prerequisite for a successful commercialization
of LoaC applications.
In this contribution, we present an advantageous
concept for pre-storage and release of liquid reagents in
LoaC cartridges. The concept is introduced for the
centrifugal microfluidic platform  but can also be
applied to pressure driven platforms. Key requirement
was to use a platform inherent parameter to control reagent
release from the storage reservoir, the hydrostatic pressure.
Liquid reagents are pre-stored in storage reservoirs
located close to the centre of rotation on a disk substrate.
The reservoir is sealed and separated by a thin bar from the
rest of the fluidic system (see Figure 1, weak-bonded
interface). At a given frequency, the centrifugal force
induces a hydrostatic pressure onto the liquid plug
sufficient for delamination and deformation of the sealing
foil at a predetermined area. The force F, exerted from the
liquid plug on the sealing foil, is given by equation (1).
with A being the impact area,
the angular frequency,
the liquid density, Δr the length and
the mean radial
position of the liquid plug. The centrifugally induced
pressure at the weak-bonded interface only depends on the
radial length on the bordering liquid plug, not on its cross-
section (hydrostatic paradox).
Via the opened fluidic path, liquids are released into
the fluidic system.
978-1-4244-4193-8/09/$25.00 ©2009 IEEE Transducers 2009, Denver, CO, USA, June 21-25, 20091991
Figure 1: Basic release mechanism: At high rotational
frequencies fc, the sealing foil at the weak-bonded interface
(represented by a thin bar) is delaminated. The pre-stored liquid
from a storage reservoir (up left) is centrifugally displaced into
the fluidic system of a Lab-on-a-Chip substrate via the opened
Figure 2: Schematic drawing of a disk-sector featuring a liquid
reagent storage structure. (a) storage reservoir, containing
100 µL of H20; (b) storage channel in a variable length, width =
400 µm, depth = 800 µm; (c) weak-bonded interface due to a
small contact area between the sealing foil and the substrate; (d)
siphon shaped release channel in a variable length, width =
400 µm, depth = 800 µm; (e) receiving reservoir which features
a venting hole.
DESIGN AND FABRICATION
Defined liquid volumes are dispensed into a
monolithic storage reservoir (a) featuring a total volume
of 170 µL at a reservoir depth of 3.5 mm. (See Figure 2
for all indicated characters). A long radial storage channel
(b) connects the storage reservoir with the weak-bonded
interface (c), shaped like a groove and tongue structure.
This design ensures a large impact area for the centrifugal
pressure, leading to a large de-lamination force (see
Equation 1), and minimizing the fluidic resistance after
delamination. The released liquid volume is centrifugally
displaced through the release channel (d) into the
receiving reservoir (e), which features at total volume of
194 µL at a reservoir depth of 3.5 mm. For assay
processing, the released liquids are guided by the siphon-
similar release channel into the receiving reservoir located
close to the centre of rotation. This allows consecutive
centrifugal microfluidic unit operations since the outer
border of the disk is not yet reached.
Fluidic structures are micro-milled into a 4-mm-thick
Cyclic Olefin Copolymer (COC) disk. Storage units are
manually sealed by an adhesive tape (Adhesive PCR Film,
AB-0558, Thermo-Scientific, Karlsruhe, Germany).
RESULTS AND DISCUSSION
Six 100 µL aliquots of water are sealed into storage
reservoirs. Release times (trel) are taken by a stopwatch.
In the first experiment, trel is determined with respect to
different radial positions of the weak-bonded interface at
two constant parameters: rotational frequency fc = 60 Hz
and width of the weak-bonded interface = 200 µm. Radial
positions are set to six values between 35 mm and 55 mm
resulting in hydrostatic pressures of 0.76, 1.00, 1.20, 1.30,
1.70, and 2.00 bar, respectively. Results are depicted in
Figure 3. Note that pressures ≤ 1.0 bar turned out to be
insufficient for delamination.
1,0 1,2 1,4 1,6 1,8 2,0
Time to release [s]
Figure 3: Time-to-release measured as a function of the
hydrostatic pressure (set by different radial position of the weak-
bonded interface) affecting the sealing foil over the weak-
bonded interface. Time-to-release trel measures the time between
starting of rotation and the point when the first droplet is
received in the receiving reservoir (6 experiments each).
The influence of the width of the weak-bonded
interface on trel is determined while keeping two
parameters constant: rotational frequency fc= 60 Hz and
radial position of the interface = 50 mm resulting in a
hydrostatic pressure of 1.70 bar. The width of the bar is
realized in 200, 300 and 400 µm. Results are depicted in
200 250 300 350 400
Time to release [s]
Width of bar [µm]
Figure 4: Time-to-release is measured as a function of the width
of the separating bar at the weak-bonded interface (6
The received volume is 69 µL at a CV of 5.7 % (n =
60). The remaining volume is left in the interconnecting
channels and is pulled back into the storage reservoir at
fc = 0 Hz due to a vacuum in the storage reservoir (see
The sealing foil (adhesive film) over a storage unit
containing H2O changed its optical transparency into a
milky appearance after one day, indicating a possible
release of contaminants from the sealing foil into the
Upon rotation, the centrifugal force pressurizes the
liquid volume in the storage reservoir whereby the air
volume (previously in the storage channel) is compressed.
At increasing frequency, the liquid-gaseous interface
becomes unstable and both volumes exchange . The
whole storage compartment is at atmospheric pressure still
(no negative pressure) since the total air and liquid
volumes in the storage chamber did not change. If the
hydrostatic pressure at the radial distance of the weak-
bonded interface is sufficient for delamination of the
sealing foil, a fluidic path to the receiving reservoir is
Upon further rotation, the liquid volume from the
storage reservoir flows via the previously delaminated
weak-bonded interface into the release channel and further
on into the receiving reservoir. Upon emptying, a negative
pressure in the unvented storage reservoir is induced.
Therefore, the exchange of air between the storage
reservoir and the receiving reservoir (which is vented) is
mandatory for maximal liquid release. Toggling the
frequency between fc = 0 Hz and fc = 60 Hz improves the
Figure 5: Storage unit on a centrifugal Lab-on-a-Chip cartridge
designed for pre-storing of 100 µL liquid volumes in the storage
reservoir (left). After delamination, 69 µL of the pre-stored
liquid are displaced into the receiving reservoir with a CV of 5.7
% (right). The remaining 31 µL approximately equal the volume
of the interconnecting channels. The center of rotation is above
CONCLUSIONS AND OUTLOOK
This work describes and investigates an advantageous
approach for the pre-storage and release of liquid reagents
in disposable Lab-on-a-Chip cartridges. Process protocols
proposed for state-of the art microfluidic platforms often
include the manual insertion of liquid reagents before or
during assay processing. Disadvantages result from the
facts that trained personnel is needed to transfer this
defined amounts of liquid. The solution is to pre-store the
liquids within the cartridge, thus enabling fully automated
analysis, allowing save operation, and reducing human
Advantage of the presented solution is its simplicity,
requiring no expensive storage vessel together with the
possibility to influence the release time by simply varying
storage structure geometries and rotational frequencies.
Most important, no additional actuator must be
implemented into an analyzer device because reagents are
released upon the instrument’s intrinsic rotational force.
Furthermore, the introduced single to use valve is time-
controllable, vapour tight and independent on wetting
properties of liquids. The concept can be easily
incorporated into cartridges for centrifugally operated Lab-
on-a-Chip platforms, being an essential building block for
fully integrated PoC-diagnostic systems.
Two specific shortcomings are related to this pre-
storage approach. First, in any case a liquid volume in the
magnitude of the interconnecting channel volume is lost.
With the described design, the loss is 30 µL ± 4 µL out of
100 µL. This amount could be reduced by lowering the
cross-section of respective channels. Secondly, chemical
stability and integrity of the pre-stored medium
encapsulated within the cartridge is still to be investigated.
A general problem when trying to pre-store liquid
volumes within cartridges is the plastic’s permeability for
vapour. This may pose an additional problem when pre-
storing liquid and freeze-dried reagents within the same
cartridge, since vapour escaping from storage reservoirs
could shorten the shelf-life of freeze-dried substances.
Hence composite layer films with low vapour
transmission rate will be preferred materials for
implementation of the introduced technology. Additional,
an appropriate bonding process for such materials has to
be implemented to guarantee a reproducible delamination
strength of the burstable seal.
 P. Yager, T. Edwards, E. Fu, K. Helton, K. Nelson,
M.R. Tam, B.H. Weigl, “Microfluidic Diagnostic
Technologies for Global Public Health”, Nature, vol.
442, pp. 412-418, 2006.
 J. Ducrée, S. Haeberle, S. Lutz, S. Pausch, F. von
Stetten, R. Zengerle: The Centrifugal Microfluidic
Bio-Disk Platform. J. Micromech. Microeng., 17
 Y. Huang, E.L. Mather, J.L. Bell, M. Madou,
“MEMS-Based Sample Preparation for Molecular
Diagnostics”, Analytical and Bioanalytical Chemistry,
vol. 372, pp. 49-65, 2002.
 R. Seetharam, Y. Wada, S. Ramachandran, H. Hess, P.
Satir, “Long-Term Storage of Bionanodevices by
Freezing and Lyophilization”, Lab Chip, vol. 6, pp.
 E. Garcia, J.R. Kirkham, A.V. Hatch, K.R. Hawkins,
P. Yager, “Controlled Microfluidic Reconstitution of
Functional Protein From an Anhydrous Storage
Depot”, Lab Chip, vol. 4, pp. 78-82, 2004.
 J.B. Findlay, S.M. Atwood, L. Bergmeyer, J.
Chemelli, K. Christy, T. Cummins, W. Donish, T.
Ekeze, J. Falvo, D. Patterson, J. Puskas, J. Quenin, J.
Shah, D. Sharkey, J.W.H. Sutherland, R. Sutton, H.
Warren, J. Wellman, “Automated Closed-Vessel
System for In-Vitro Diagnostics Based on Polymerase
Chain-Reaction”, Clin. Chem., vol. 39, pp. 1927-1933,
 R. Boden, M. Lehto, J. Margell, K. Hjort, J.A.
Schweitz, “On-Chip Liquid Storage and Dispensing
for Lab-on-a-Chip Applications”, J. Micromech.
Microeng., vol. 18, pp. 7-ff, 2008.
 J.L. García-Cordero, “Low-Cost Microfluidic Single-
Use Valves And On-Board Reagent Storage Using
Laser-Printer Technology”, in Digest Tech. Papers
MEMS ’09 Conference, Sorrento, January 25-29, pp.
439 - 442.
 J. Ducrée, S. Haeberle, S. Lutz, S. Pausch, F. von
Stetten, R. Zengerle, “The Centrifugal Microfluidic
Bio-Disk Platform”, J. Micromech. Microeng., vol. 17,
pp. 103-115, 2007.
D. Mark, “Aliquoting Structure for Centrifugal
Microfluidics Based on a New Pneumatic Valve”, in
Digest Tech. Papers MEMS ’08 Conference, Tucson,
January 13-17, pp. 611-614.
We acknowledge financial support by the German
Federal Ministry of Education and Research (project
Zentrilab, link number V3BIO007):
* Felix von Stetten, tel: +49-761-203-7393; fax: +49-
761-203-7539; email: email@example.com