Hh

JLa

b

c

bd

e

a

ARRAA

KCMFHS

1

mLpp

tptea

baftWa

0h

Sensors and Actuators B 190 (2014) 451– 458

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical

journa l h om epage: www.elsev ier .com/ locat e/snb

ighly integrated COP monolithic membrane microvalves by robustot embossing

aione Etxebarriaa,b,∗, Javier Berganzoa,b, Jorge Elizaldea,b,uis José Fernándezc,d,e, Aitor Ezkerraa,b

CIC microGUNE, Polo de innovación Garaia. Goiru 9, 20500 Arrasate-Mondragón, SpainMicrosystems Department, IK4-Ikerlan, Polo de innovación Garaia, Goiru 9, 20500 Arrasate-Mondragón, SpainGroup of Structural Mechanics and Materials Modelling (GEMM), Centro Investigación Biomédica en Red, Bioingeniería,iomateriales y nanomedicina (CIBER-BBN), Zaragoza, SpainAragon Institute of Engineering Research (I3A), University of Zaragoza, Zaragoza, SpainAragon Institute of Biomedical Research, Instituto de Salud Carlos III, Aragon, Spain

r t i c l e i n f o

rticle history:eceived 23 May 2013eceived in revised form 9 August 2013ccepted 30 August 2013vailable online 8 September 2013

a b s t r a c t

This paper presents an in-plane pneumatically actuated membrane-type microvalve, entirely made ofCyclic Olefin Polymer (COP). The body of the valve is fabricated following a robust hot-embossing methodwith SU-8 master moulds, producing devices with repetitive dimensions at wafer-level. Sealing is per-formed by applying a suitable solvent on the COP membrane, rendering monolithic devices, free fromassembly errors. Various design parameters have been studied to obtain different working regimes, with

eywords:OPicrovalve

low controlot embossing

maximum flow rates of 8.5 ml/min being successfully regulated and fully stopped. Owing to its fabricationmethod and characteristics, these devices represent a reliable and low-cost solution for the integrationof microfluidic control in mass-produced lab-on-a-chip devices.

© 2013 Elsevier B.V. All rights reserved.

olvent-bonding

. Introduction

Microvalves are critical for fluid flow control and thus, one of theost important components for the realization of fully integrated

ab-on-a-Chip (LoC) systems [1–5]. High potential applications forrecise volume handling include chemical analysis, gas/liquid sam-le injection, flow regulation and gas or liquid sealing [6].

Microvalves are basically categorized into active and passive,he former being best suited for autonomous LoCs. Actuationrinciples include pneumatic [7–12], thermopneumatic [13–17],hermomechanical, piezoelectric [18–21], electrostatic [22–26],lectromagnetic [27–32], electrochemical and chemical [33–39]nd capillary force [40–43].

Amongst these, pneumatically actuated membrane valves haveeen successfully used in many applications due to their low costnd simple fabrication. A variety of devices has been demonstratedor silicon [44–46] and glass–silicon [47,48] as well as for less tradi-

ional materials such as elastomers [49–52] and polymers [53–56].

ithin the latter, cyclic olefin polymer (COP) has recently emergeds an attractive material [57,58] due to its high optical clarity

∗ Corresponding author. Tel.: +34 943710212; fax: +34 943710212.E-mail address: jetxebarria@ikerlan.es (J. Etxebarria).

925-4005/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.snb.2013.08.103

(into deep-UV range) [59], high bio-compatibility [60,61], low auto-fluorescence, low water absorption [62–65] and good chemicalresistance, also against organic solvents [58,62,63].

Although valves represent one of the most researched topicsupon microfluidics over the past 30 years, there is not yet any exam-ple that can be truly integrated in complex LoC systems, which isrobust, simple, versatile, low-cost and free from assembly errors.In this paper, a simple and monolithic active valve is presented,showing wide linear flow regulation capabilities and low closingpressure. The valves have been fabricated by a robust hot emboss-ing protocol, showing industrial-quality dimensional control; itscombination with COP provides low-cost and ready integrationat laboratory level that is compatible with mass-production.Finally, a reliable solvent-based bonding method is introduced,which renders planar, strong sealing of the actuation membrane,allowing control of micrometric clearances for valve performanceenhancement. Hence, a highly versatile valve concept and a directprototype-to-mass-scale fabrication method are presented, whichcan be highly useful for the LoC research community and industry.

2. Device architecture and working principle

An image of the developed active-type microvalve is presentedin Fig. 1. As it can be observed, two different heights are defined,

452 J. Etxebarria et al. / Sensors and Actuators B 190 (2014) 451– 458

Fig. 1. Image of the finished microvalve in a 1 cm2 dye, filled with a blue-coloredsolution, which highlights the different parts of the structure. (For interpretation ofto

Fdvirsagctta

rm[potpba

Fig. 3. Theoretical variation of the valve’s flow profile, for a given valve seat (0.7 mm)

Fc

he references to color in this figure legend, the reader is referred to the web versionf this article.)

ig. 2a. On one hand, the valve chamber is delimited by the outeriameter (D) and the clearance between the membrane and thealve seat (gap or G). On the other hand, the valve seat is delim-ted by the inner diameter (C). The membrane that seals the deviceegulates the flow through it by deflecting and reducing the fluidicection under the action of an external pressure source. Thus, uponpplication of pressure, the membrane deflects and the fluidic pathets narrower, consequently reducing the flow rate. When suffi-ient pressure is applied, Fig. 2b, the membrane collapses againsthe seat, stopping the flow and closing the microvalve. Owing tohis construction, the valve can achieve extensive flow modulationnd close at relatively low pressure.

In general terms, the valve’s flow rate profile depends on theate of increase of fluidic resistance per unit membrane displace-ent. This can be modelled using total pressure drop equations

66], which provide an estimate of the pressure drop across eachart of the device (microfluidic channel and valve) as a functionf geometry. According to these, for a given seat diameter the ini-ial gap size will influence the maximum flow rate, the percentage

ressure drop across the valve and its rate of increase as the mem-rane deflects. Hence, if the gap is initially large, the pressure dropt the valve will be small for low values of membrane deflection,

ig. 2. (a) Schematic drawing and cross-section of the normally-open microvalve, withoulosed state.

and valve diameter (2.5 mm), as the gap size is reduced (� = 40 �m, � = 20 �m,� = 10 �m).

giving a parabolic flow rate decay. Past a certain level of deflec-tion, the resistance at the valve will become dominant, resulting ina linearly decaying flow rate, as predicted by Poiseuille’s equation[67]. For smaller gap sizes, the valve dominates over a wider range,resulting in a rather linearly controllable flow rate. Fig. 3 depictssuch behaviour, as predicted by the equations. With respect to theseat diameter, its effect is easier to predict, simply resulting in a pro-portional change in closing pressure due to a change in the value ofmembrane deflection per unit actuation pressure [68]. Hence, fromthe previous, we can select the maximum flow rate and its regu-lation profile through the gap size and the pressure range throughthe seat diameter.

The effect of these design parameters on the valve’s behaviourhas been demonstrated through a design array, having a diame-ter of 2.5 mm, two different seat diameters (0.7 mm and 1.0 mm)and three gap heights (10 �m, 20 �m, 40 �m). In order to real-ize these architectures, a robust membrane bonding technique,which allows narrow gaps without collapse and a highly repet-itive hot-embossing method have been developed. The resulting

valves feature a low foot-print, monolithic construction withexcellent dimensional control and neatly defined micro-sizedfeatures.

t the diaphragm. (b) Cross-section of a COP microvalve with a COP diaphragm in its

J. Etxebarria et al. / Sensors and Actuators B 190 (2014) 451– 458 453

ter mo

3

81t

3

sp

sp

taru3tteFpioi

3

tcst7cia

Fig. 4. Schematic of mas

. Fabrication procedure

The fabrication steps include manufacturing of the mould in SU- and its subsequent embossing into a COP substrate (Zeonor®420R), followed by solvent bonding of a COP membrane to sealhe microchannel and finalize the valve.

.1. Master mould fabrication

The fabrication protocol of the polymeric SU-8 master mould, ashown in Fig. 4, is based on a previously developed technique forrototyping SU-8 lab-on-a-chip devices [69].

The manufacturing process for each layer of the mouldedtructure consists of spin coating, soft baking, UV exposure andost-exposure baking of the SU-8 photoresist.

First, the valve chamber layer, whose thickness corresponds tohe gap, is deposited by spin coating on a100 mm pyrex substratend soft baked at 90 ◦C for 7 min, using a progressive temperatureamp to avoid microcracking (Fig. 4i). The layer is later exposednder UV with a dosage of 130 mJ/cm2 and post baked at 90 ◦C for

min, again with progressive temperature ramps (Fig. 4ii). Next,he valve seat layer (80 �m) is spun and processed in a similar way,he soft baking being at 90 ◦C for 15 min (Fig. 4iii), followed by a UVxposure of 200 mJ/cm2 and a post bake at 90 ◦C for 3 min (Fig. 4iv).inally, the valve’s outlet layer is spun (Fig. 4v) and processed as therevious layer (Fig. 4vi), the wafer being subsequently developed

n a Propylen Glycol Monomethyl Ether Acetate (PGMEA) bath tobtain the desired mould structure (Fig. 4vii). The fabricated moulds depicted in Fig. 5.

.2. Hot embossing protocol

The procedure consists in the replication of the valve struc-ure by pressing the mould against a COP substrate, under fixedonditions of pressure and temperature. Here, the mould and theubstrate are heated up to 157 ◦C under vacuum and pressedogether with an increasing force of 2.5 kN/min up to a maximum of

.5 kN. Heating is maintained for 80 min, upon which the parts areooled down to the de-moulding temperature (Td) while maintain-ng the compression force. Finally at Td, both parts are separatednd allowed to cool down to room temperature.

uld fabrication process.

It should be noted that defects may appear on the replicatedpart and that most of these do not result from failures in fillingthe mould, but during de-moulding, where the forces exerted toseparate the parts are the main cause [70–73], i.e. friction force,augmenting with the difference in coefficient of thermal expan-sion (CTE) between the mould and the COP and adhesive forces atall contact surfaces. Minimization of these forces to obtain defect-free pieces required a careful choice of the mould material and theTd.

In terms of mould material, SU-8 was chosen due to its rigidity,preventing deformation or fracture during replication, and its Tg,considerably higher than that of COP. With respect to Td, an opti-mization protocol has been followed to find the point where thecombination of stress at the interface and the prevailing mechan-ical strength of the substrate allow defect-free de-moulding,without causing re-flow (Td too high) or mounds (Td too low)[74,75].

As a general rule, the embossing should take place at 15 ◦C to20 ◦C above Tg and de-moulding at 10 ◦C below, but this has to beempirically optimized for each material [75]. Initial embossing testshave been performed at a previously reported Td of 120 ◦C for COP[76].

3.3. Bonding

Once the structures have been defined the membrane has to bebonded in order to finalize the valve. Current COP bonding meth-ods have limitations, including channel blockage or distortion, poorbonding strength, low durability and inhomogeneous channel sur-face properties [77].

Solvent bonding, a common technique used for joining mouldedparts of amorphous thermoplastic, has been chosen due to itssimplicity, low-cost and excellent bonding strength. The solventdissolves the surface of the two parts allowing the polymerchains to intertwine. Once the solvent evaporates, the chainslose their mobility and a monolithic COP-to-COP bond is created.This requires that the solvent be strong enough to release the

chains without degrading the polymer and evaporate slowly toease manipulation of the membrane. In this work, chlorobenzenehas been chosen, as it meets the requirements for a successfulbonding.

454 J. Etxebarria et al. / Sensors and Actuators B 190 (2014) 451– 458

Fig. 5. (a) SU-8 master mould with different valve designs for hot embossing. (b) Close-up view of one valve mould structure (1 × 1 cm2).

Expe

tNsu

4

chit

mtama

dttadab

va

to a PMMA packaging that houses the microvalve (Microliq-uid, Spain), which allows fast replacement of devices forleak-free operation. The setup and packaging are shown inFigs. 6 and 7.

Fig. 6. (a) Block diagram of the experimental set-up. (b)

The process starts by spinning 4 ml of chlorobenzene on top ofhe membrane until a dry and slightly adhering surface is obtained.ext, the membrane is brought into contact with the patterned

ubstrate and pressed against it. Finally, the devices are degassednder vacuum to ensure any trace of solvent is removed.

. Characterization and Setup

The characterization protocol is intended to identify the valve’sapabilities for flow regulation and blockage; thus, performanceas been evaluated under flow conditions likely to be encountered

n LoC devices. A constant pressure drop of 1 bar is imposed acrosshe device with no actuation.

In terms of actuation, any element capable of deflecting theembrane could be employed, which includes a good number of

hose mentioned in the introduction. However, pneumatic actu-tion was chosen because it is a widely available and affordableeans to extract the full potential of the valve, offering uniform

nd controllable force over the membrane.As the pneumatic load on the membrane is increased, its

eflection towards the seat increases the fluidic resistance ofhe system and provides flow regulation. When the pressure onhe membrane is sufficiently high, this collapses against the seatnd blocks the flow. The linearity of the load vs. flow rate willetermine the quality of the valve as a flow regulator whereas

low value of pressure at collapse will determine that for fluid

lockage.

The setup includes a positive pressure flow controller that pro-ides constant pressure drop (Fluigent MFCSTM FLEX, France),

flow microsensor to measure the flow rate passing through

rimental set-up for the characterization of microvalves.

the microvalve (Sensirion CMOSens©, Switzerland) and a pres-sure controller (EFD1500XL, USA) to regulate the pneumaticactuation of the membrane. These elements are connected

Fig. 7. PMMA packaging showing the microvalve in place and the fluidic and pneu-matic ports required for actuation.

J. Etxebarria et al. / Sensors and Actuators B 190 (2014) 451– 458 455

Fig. 8. Confocal Topography of a replica obtained when de-moulding under Tg . Themound-like defect, which can be observed at the edge of the inner diameter, isa

5

5

rmaofaod

iloaofs

5.2. Solvent bonding

Fcs

round 18 �m on the valve seat, which is almost the half of the gap height (40 �m).

. Results and discussion

.1. Hot embossing

Initial embossing test was performed at a Td of 120 ◦C for COP,esulting in mound type deformations up to 18 �m high, asym-etrically arranged at the edges of the valve seat, Fig. 8. Mounds

re caused by shear stresses higher than the yield stress of COP,riginating from an excess friction at the mould/replicate inter-ace. Friction is dependent on the contraction of the thermoplasticgainst the mould due to CTE mismatch [70,73,78,79]. The presencef such defects was not previously reported at this Td [75], probablyue to the low contact surface of the channels.

Alternatively, higher temperatures have been tested, consider-ng that closer to the Tg, the contraction of the material will beower, reducing the aforementioned shear stress and the formationf mounds. Furthermore, at higher Td the polymer is still reflowingnd the pattern recovery above this [75,80] can reduce the severity

f the defects. In order to find the optimum, Td has been increasedrom the initial 120 ◦C and the resulting mound measured, usingurface profilometry (Veeco Dektak150, USA).

ig. 10. SEM images of a replica obtained employing the optimized hot embossing procesollapse. (a) Image of the whole microvalve structure with a gap of 10 �m; (b) close-up ieat and the membrane.

Fig. 9. Confocal Topography of a replica obtained employing the optimized hotembossing process. No mounds are generated due to the equilibrium betweenmound generation and recovery of pattern.

Measurements confirm that as Td increases, the resultingmounds are reduced, becoming negligible (<1 �m) at a Td of 143 ◦C,Fig. 9.

De-moulding at Td ≥ Tg, also prevents non-uniform shrinking ofthe substrate during cooling, which would affect the dimensionalcontrol of the devices and its repeatability. Discussions around thistype of defects have not been found in the literature, despite being aknown problem, particularly for those works where an elastomericmould is employed. The valves resulting from the aforementionedprotocol have been measured and compared at chip and waferlevel in a high-precision automatic dicing machine (Disco DAD321,Japan). It has been found that the total contraction is 0.67 ± 0.036%,indicative of a large uniformity and very close to the specifica-tions of the manufacturer (Zeonor®, Japan) (0.5–0.7%) for injectionmoulding. Therefore, a defect-free process has been developed, pro-viding repeatability and a dimensional control comparable to thatof standard mass-production methods.

This solvent bonding method allows fabrication of microvalveswith a minimum distance of 10 �m between the valve seat and the

s and the adequate solvent bonding technique, which allows narrow gaps withoutmage of the gap and the suspended membrane, showing no collapse between the

456 J. Etxebarria et al. / Sensors and Actuators B 190 (2014) 451– 458

F 2.5 mt 20 �mi this fi

mpma0f

5

sFbtpmcnt

bstwtl

trpf

6

wibmsrpt

ig. 11. Plots of flow rate vs. membrane pressure for a given microvalve diameter (he theoretical prediction can be appreciated for reducing gap sizes (� = 40 �m, � =

ncrease that limits flow regulation. (For interpretation of the references to color in

embrane (gap), Fig. 10a and b. Furthermore, unimpaired opticalroperties and fast processing are achieved. The deformation of theembrane at the centre as a result of bonding pressure is remark-

bly small, being 0.04% (±0.013%) for the a 2.5 mm Ø membrane,.024% (±0.014%) for the 3 mm Ø membrane and 0.016% (±0.011%)or the 4 mm Ø membrane.

.3. Characterization

The array of gaps and seat diameters employed highlights theeparate effects of the valve’s design features on performance.ig. 11 shows plots of flow rate vs. pressure for each design. As it cane seen, the flow profile dependence with the gap agrees well withhe theory, as shown by the solid line, except around the closingressure, which is substantially higher than predicted. In fact, theodel assumes that the flow is interrupted when the membrane

ontacts the valve seat. However, this is not true for rigid compo-ents like those of the valve, thus extra pressure is needed to forcehe membrane sufficiently against the seat, as observed.

The effect of an increasing valve seat also follows the theoreticalehaviour, resulting in larger flow for similar membrane pressures,ince the value of deflection, and thus closure, is lower as we moveowards the edges of the membrane. In principle, this implies aider regulation range. However, in practice this difference is seen

o reduce significantly as the gap decreases, which sets the lowerimit for the flow tailoring capabilities predicted by the theory.

Therefore, we prove that the present architecture can be usedo construct valves with good flow linearity over a wide actuationange on the membrane. Furthermore, the devices have excellentressure drop characteristics, with tunable flow rates well suitedor LoC applications.

. Conclusions

In this work, a microvalve architecture has been presented,hereby two dimensions can be easily modified to tune the max-

mum flow rate and the flow regulation profile. i.e., the spaceetween the membrane and the valve seat (gap) can be used toodify the flow rate and linearize the flow profile, and the valve

eat can be used to modify the range of membrane pressuresequired for regulation. The array of devices employed shows thatast a certain gap size, the flow turns from parabolic to linear andhat the resulting effect on the closing pressure can be tuned by

m) and (a) 0.7 mm seat diameter or (b) 1.0 mm seat diameter. The agreement with, � = 10 �m), although the rigidity of the material limits imposes a closing pressuregure legend, the reader is referred to the web version of this article.)

the valve seat diameter. The pressure drop characteristics of thedevices allow flow rate values well suited for LoC applications.

Furthermore, this microvalve provides improved performancefor liquid flow regulation, in terms of its ability to provide low flowrates suitable for LoC applications, ranging from 200 nl/min up to9 ml/min, in comparison with other works with flow rates as high as1.3 l/min [30], ability to fully close the valve with negligible leakageof 0.2 �l/min, in comparison with other works that present non-negligible leakage of 3.9 �l/min [81] or a considerable leakage of420 �l/min [82], improved repeatability [83], and faster responsetime [84].

The devices have been fabricated using a simple and low-costhot embossing method, which renders virtually defect-free deviceswith uniform contraction at all points of the replica, achieving val-ues similar to those of industrial processes like injection moulding(0.67%). The devices are finished off by solvent-bonding the seal-ing membrane, resulting in a monolithic component with excellentflatness at replica level and allowing the fabrication of valves withuniform gaps as low as 10 �m.

The useful characteristics of the devices presented in this workand the quality of their fabrication method evidence the possibilityto prototype highly performing, integrated fluidic control devicesin rigid polymers at low cost, which could be highly beneficial forthe Lab-on-a-chip community.

Summarizing, this approach solves three important problems.First, the simple and monolithic active valve introduced in this workrepresents a real alternative for microfluidic control, as it showstunable and highly repetitive flow regulation capabilities, allowingto cater for varying flow control needs with a single valve conceptand with high performance standards. Second, the valve shows evi-dence of its enormous integration potential, owing to its monolithicconstruction in ordinary materials, which can simplify the complextask of embedding microfluidic control in LoCs. Third, the valve is areal alternative for mass-production, as demonstrated by the excel-lent fabrication results obtained with one of the materials of choicefor mass-produced LoCs, which could be readily replicated in anyother thermosetting polymer.

Acknowledgements

The authors would like to acknowledge Mikel Gómez frommicroGUNE/IK4-Ikerlan for technical support.

d Actu

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J. Etxebarria et al. / Sensors an

eferences

[1] P.A. Auroux, D. Iossifidis, D.R. Reyes, A. Manz, Micro total analysis systems.2. Analytical standard operations and applications, Analytical Chemistry 74(2002) 2637–2652.

[2] P.S. Dittrich, K. Tachikawa, A. Manz, Micro total analysis systems. Latestadvancements and trends, Analytical Chemistry 78 (2006) 3887–3908.

[3] A. Manz, N. Graber, H.M. Widmer, Miniaturized total chemical analysis systems:a novel concept for chemical sensing, Sensors and Actuators B: Chemical 1(1990) 244–248.

[4] D.R. Reyes, D. Iossifidis, P.A. Auroux, A. Manz, Micro total analysis sys-tems. 1. Introduction, theory, and technology, Analytical Chemistry 74 (2002)2623–2636.

[5] T. Vilkner, D. Janasek, A. Manz, Micro total analysis systems: recent develop-ments, Analytical Chemistry 76 (2004) 3373–3385.

[6] K.W. Oh, C.H. Ahn, A review of microvalves, Journal of Micromechanics andMicroengineering 16 (2006) R13–R39.

[7] P.K. Yuen, L.J. Kricka, P. Wilding, Semi-disposable microvalves for use withmicrofabricated devices or microchips, Journal of Micromechanics and Micro-engineering 10 (2000) 401–409.

[8] Y.C. Wang, M.H. Choi, J. Han, Two-dimensional protein separation withadvanced sample and buffer isolation using microfluidic valves, AnalyticalChemistry 76 (2004) 4426–4431.

[9] A. Luque, J.M. Quero, C. Hibert, P. Flückiger, A.M. Gana-Calvo, Integrable siliconmicrofluidic valve with pneumatic actuation, Sensors and Actuators A: Physical118 (2005) 144–151.

10] M. Kanai, H. Abe, T. Munaka, Y. Fujiyama, D. Uchida, A. Yamayoshi, et al., Microchamber for cellar analysis integrated with negligible dead volume sampleinjector, Sensors and Actuators A: Physical 114 (2004) 129–134.

11] Z. Hua, Y. Xia, O. Srivannavit, J.M. Ruillard, X. Zhou, X. Gao, et al., A versa-tile microreactor platform featuring a chemical-resistant microvalve array foraddressable multiplex syntheses and assays, Journal of Micromechanics andMicroengineering 16 (2006) 1433–1443.

12] J.Y. Baek, J.Y. Park, J.I. Ju, T.S. Lee, S.H. Lee, A pneumatically controllable flexibleand polymeric microfluidic valve fabricated via in situ development, Journal ofMicromechanics and Microengineering 15 (2005) 1015–1020.

13] X. Yang, C. Grosjean, Y.C. Tai, Design, fabrication, and testing of micromachinedsilicone rubber membrane valves, Journal of Microelectromechanical Systems8 (1999).

14] H. Takao, M. Ishida, Microfluidic integrated circuits for signal processing usinganalogous relationship between pneumatic microvalve and MOSFET, Journalof Microelectromechanical Systems 12 (2003).

15] C.A. Rich, K.D. Wise, A high-flow thermopneumatic microvalve with improvedefficiency and integrated state sensing, Journal of Microelectromechanical Sys-tems 12 (2003) 201–208.

16] J.H. Kim, K.H. Na, C.J. Kang, D. Jeon, Y.S. Kim, A disposable thermopneumatic-actuated microvalve stacked with PDMS layers and ITO-coated glass,Microelectronic Engineering 73–74 (2004) 864–869.

17] D. Baechi, R. Buser, J. Dual, A high density microchannel network with inte-grated valves and photodiodes, Sensors and Actuators A: Physical 95 (2002)77–83.

18] P. shao, Z. Rummler, W.K. Schomburg, Polymer micro piezo valve with a smalldead volume, Journal of Micromechanics and Microengineering 14 (2004)305–309.

19] H.Q. Li, D.C. Roberts, J.L. Steyn, K.T. Turner, O. Yagliolu, N.W. Hagood, et al.,Fabrication of a high frequency piezoelectric microvalve, Sensors and ActuatorsA: Physical 111 (2004) 51–56.

20] I. Chakraborty, W.C. Tang, D.P. Bame, T.K. Tang, MEMS micro-valve for spaceapplications, Sensors and Actuators A: Physical 83 (2000) 188–193.

21] C. Lee, E.H. Yang, S.M. Saeidi, J.M. Khodadadi, Fabrication characterization andcomputational modeling of a piezoelectrically actuated microvalve for liquidflow control, Journal of Microelectromechanical Systems 15 (2006) 686–696.

22] M. Shikida, K. Sato, S. Tanaka, Y. Kawamura, Y. Fujisaki, Electrostatically drivengas valve with high conductance, Journal of Microelectromechanical Systems3 (1994) 76–80.

23] C. Goll, W. Bacher, B. Bustgens, D. Mass, R. Ruprecht, W.K. Schomburg, Anelectrostatically actuated polymer microvalve equipped with a movable mem-brane electrode, Journal of Micromechanics and Microengineering 7 (1997)224–226.

24] J.K. Robertson, K.D. Wise, A low pressure micromachined flow modulator, Sen-sors and Actuators A: Physical 71 (1998) 98–106.

25] W. Van der Wijngaart, H. Ask, P. Enoksson, G. Stemme, A high-stroke, high-pressure electrostatic actuator for valve applications, Sensors and Actuators A:Physical 100 (2002) 264–2671.

26] M.M. Teymoori, E. Abbaspour-Sani, Design and simulation of a novel electro-static peristaltic micromachined pump for drug delivery applications, Sensorsand Actuators A: Physical 117 (2005) 222–229.

27] B. Bae, H. Kee, S. Kim, Y. Lee, T. Sim, Y. Kim, et al., In vitro experiment of thepressure regulating valve for a glaucoma implant, Journal of Micromechanicsand Microengineering 13 (2003) 613–619.

28] S. Bohm, G.J. Burger, M.T. Korthorst, F. Roseboom, A micromachined silicon

valve driven by a miniature bi-stable electro-magnetic actuator, Sensors andActuators A: Physical 80 (2000) 77–83.

29] C. Fu, Z. Rummle, W. Chomburg, Magnetically driven micro ball valves fab-ricated by multilayer adhesive film bonding, Journal of Micromechanics andMicroengineering 13 (2003) 96–102.

[

[

ators B 190 (2014) 451– 458 457

30] K.W. Oh, R. Rong, C.H. Anh, In-line micro ball valve through polymer tubing,microTAS (2001) 407–408.

31] D. De Bhailís, C. Murray, M. Duffy, J. Alderman, G. Kelly, S.C.O. Mathúna, Mod-elling and analysis of a magnetic microactuator, Sensors and Actuators A:Physical 81 (2000) 285–289.

32] J. Sutanto, P.J. Hesketh, Y. Berthelot, Design, microfabrication and testingof a CMOS compatible bistable electromagnetic microvalve with latch-ing/unlatching mechanism on a single wafer, Journal of Micromechanics andMicroengineering 16 (2006) 266–275.

33] C.R. Neagu, J.G.E. Gardeniers, M. Elwenspoek, J.J. Kelly, An electrochemicalmicroactuator: principle and first results, Journal of MicroelectromechanicalSystems 5 (1996) 2–9.

34] H. Suzuki, R. Toneyama, Integrated microfluidic system with electrochemicallyactuated on-chip pumps and valves, Sensors and Actuators B: Chemical 96(2003) 38–45.

35] K. Takahashi, K. Yoshino, S. Hatano, K. Nagayama, T. Asano, Novel appli-cations of thermally controlled microbubble driving system, IEEE (2001)286–289.

36] D.T. Eddington, D.J. Beebe, A valved responsive hydrogel microdispensingdevice with integrated pressure source, Journal of Microelectromechanical Sys-tems 13 (2004) 586–593.

37] A. Richter, D. Kuckling, S. Howitz, T. Gehring, K.-F. Arndt, Electronicallycontrollable microvalves based on smart hydrogels: magnitudes and poten-tial applications, Journal of Microelectromechanical Systems 12 (2003)748–753.

38] L.-M. Low, S. Seetharaman, K.-Q. He, M.J. Madou, Microactuators towardmicrovalves for responsive controlled drug delivery, Sensors and Actuators B:Chemical 67 (2000) 149–160.

39] C. Yamahata, M. Chastellain, V.K. Parashar, A. Petri, H. Hoffmann, M.A.M. Gijs,Plastic micropump with ferrofluidic actuation, Journal of Microelectromechan-ical Systems 14 (2005) 96–102.

40] J. Lee, H.S. Moon, J. Fowler, T. Schoellhammer, C.-J. Kim, Electrowetting andelectrowetting-on-dielectric for microscale liquid handling, Sensors and Actu-ators A: Physical 95 (2002) 259–268.

41] P. Paik, V.K. Pamula, M.G. Pollack, R.B. Fair, Electrowetting-based droplet mixersfor microfluidic systems, Lab on a Chip 3 (2003) 28–33.

42] P.R.C. Gascoyne, J.V. Vykoukal, J.A. Schwartz, T.J. Anderson, D.M. Vykoukal, K.W.Current, et al., Dielectrophoresis-based programmable fluidic processors, Labon a Chip 4 (2004) 299–309.

43] V. Srinivasan, V.K. Pamula, R.B. Fair, An integrated digital microfluidic lab-on-a-chip for clinical diagnostics on human physiological fluids, Lab on a Chip 4(2004) 310–315.

44] L. Bousse, E. Dijkstra, O. Guenat, High-density arrays of valves and interconnectsfor liquid switching, in: Solid-State Sensors and Actuators Workshop, HiltonHead, SC, 1996, pp. 272–275.

45] T. Ohori, S. Shoji, K. Miura, A. Yotsumoto, Partly disposable three-waymicrovalve for a medical micro total analysis system (TAS), Sensors and Actu-ators A: Physical 64 (1998) 57–62.

46] C. Vieider, O. Ohman, H. Elderstig, A penumatically actuated micro valve witha silicone rubber membrane for integration with fluid-handling systems, in:Proceedings of IEEE International Conference Solid-State Sensors and Actua-tors, vol. 2, Stockholm, Sweden, 1995, pp. 284–286.

47] D. Baechi, R. Buser, J. Dual, A high density microchannel network with inte-grated valves and photodiodes, Sensors and Actuators A: Physical 95 (2002)77–83.

48] X. Yang, C. Grosjean, Y.C. Tai, C.M. Ho, A MEMS thermopneumatic siliconerubber membrane valve, Sensors and Actuators A: Physical 64 (1998) 101–108.

49] K. Hosokawa, R. Maeda, A pneumatically-actuated three-way microvalve fab-ricated with polydimethylsiloxane using the membrane transfer technique,Journal of Micromechanics and Microengineering 10 (2000), 514-420.

50] X.Q. Ren, M. Bachman, C. Sims, G.P. Li, N. Allbritton, Electroosmotic proper-ties of microfluidic channels composed of poly-(dimethylsiloxane), Journal ofChromatography B 762 (2001) 117–125.

51] M.A. Unger, H.P. Chou, T. Thorsen, A. Scherer, S.R. Quake, Monolithic microfab-ricated valves and pumps by multilayer soft lithography, Science 288 (2000)113–116.

52] Y. Xia, G.M. Whitesides, Soft lithography, Annual Review of Materials Science28 (1998) 153–184.

53] R.C. Anderson, X. Su, G.J. Bogdan, J. Fenton, A miniature integrated device forautomated multistep genetic assays, Nucleic Acids Research 28 (2000).

54] S. Bohm, W. Olthuis, P. Bergveld, A plastic micropump constructed with conven-tional techniques and materials, Sensors and Actuators A: Physical 77 (1999)223–228.

55] W.K. Schomburg, R. Ahrens, W. Bacher, J. Martin, V. Saile, AMANDA—surfacemicromachining, molding, and diaphragm transfer, Sensors and Actuators A:Physical 76 (1999) 343–348.

56] S. Sjolander, C. Urbaniczky, Integrated fluid handling system for biomolecularinteraction analysis, Analytical Chemistry 63 (1991) 2338–2345.

57] P.S. Nunes, P.D. Ohlsson, O. Ordeig, J.P. Kutter, Cyclin olefin polymers: emerg-ing materials for lab-on-a-chip applications, Microfluidics and Nanofluidics 9(2010) 145–161.

58] C.-W. Tsao, D.L. Devoe, Bonding of thermoplastic polymer microfluidics,Microfluidics and Nanofluidics 6 (2009) 1–16.

59] M. Hansen, D. Nilsson, D.M. Johansen, S. Balslev, A. Kristensen, A nanoimprintedpolymer lab-on-a-chip with integrated optics, in: Conference: Advancementsin Polymer Optics Design, Fabrication, and Materials, vol. 5872, 2005, pp. 76–82.

4 d Actu

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

58 J. Etxebarria et al. / Sensors an

60] A. Bhattacharyya, C. Klapperich, Thermoplastic microfluidic device for on-chippurification of nucleic acids for disposable diagnostics, Analytical Chemistry 78(2006) 788–792.

61] S.H. Choi, D.S. Kim, T.H. Kwon, Microinjection molded disposable microfluidiclab-on-a-chip for efficient detection of agglutination, Microsystem Technolo-gies 15 (2009) 309–316.

62] TOPAS Advanced Polymers, Topas. Available at: http://www.topas.com/company (accessed September 2013).

63] Zeon Chemicals, Zeonex and Zeonor. Available at: http://www.zeonchemicals.com/ (accessed September 2013).

64] Japan Synthetics, Rubber, ARTON. Available at: http://www.jsr.co.jp/jsr e/(accessed September 2013).

65] Mitsui Chemicals, APEL. Available at: http://www.mitsuichemicals.com/(accessed September 2013).

66] J. Berthier, P. Silberzan, Microfluics for Biotechnology, Artech House Inc., MA,USA, 2006.

67] H. Bruus, Theoretical Microfluidics, Oxford University Press (OUP), USA, 2008.68] S.P. Timoshenko, S. Woinowsky-Krieger, Theory of Plates and Shells, second

ed., MacGraw-Hill, New York, 1959.69] M. Agirregabiria, F.J. Blanco, J. Berganzo, M.T. Arroyo, A. Fullaondo, K. May-

ora, et al., Fabrication of SU-8 multilayer microstructures based on successiveCMOS compatible adhesive bonding and releasing steps, Lab on a Chip 5 (2005)545–552.

70] M. Dirckx, H. Taylor, D. Hardt, High temperature de-molding for cycle timereduction in hot embossing, in: Proceedings Society of Plastics EngineersAnnual Technical Conference, Cincinnati, OH, 2007, pp. 2972–2976.

71] Y.H. Guo, G. Liu, X.L. Zhu, Y.C. Tian, Analysis of the demoldingforces during hot embossing, Microsystem Technologies—Micro-andNanosystems—Information Storage and Processing Systems 13 (2007)411–415.

72] Y.J. Juang, L. Lee James, K.W. Koelling, Hot embossing in microfabrication. PartII: Rheological characterization and process analysis, Polymer Engineering andScience 42 (2002) 551–566.

73] G. Yuhua, L. Gang, X. Yin, T. Yangchao, Study of the demoldingprocess—implications for thermal stress, adhesion and friction control, Journalof Micromechanics and Microengineering 17 (2007) 9.

74] Z. Song, B.H. You, J. Lee, S. Park, Study on demolding temperature in thermalimprint lithography via finite element analysis, Microsystem Technologies 14(2008) 1593–1597.

75] M.B. Esch, S. Kapur, G. Irizarry, V. Genova, Influence of master fabrication tech-niques on the characteristics of embossed microfluidic channels, Lab on a Chip3 (2003) 121–127.

76] X. Illa, O. Ordeig, D. Snakenborg, A. Romano-Rodriguez, R.G. Compton, J.P.Kutter, A cyclo olefin polymer microfluidic chip with integrated gold micro-electrodes for aqueous and non-aqueous electrochemistry, Lab on a Chip 10(2010) 1254–1261.

77] T.I. Wallow, A.M. Morales, B.A. Simmons, M.C. Hunter, K.L. Krafcik, L.A. Domeier,et al., Low-distortion, high-strength bonding of thermoplastic microfluidicdevices employing case-II difusion-mediated permeant activation, Lab on aChip 7 (2007) 1825–1831.

78] M. Dirckx, Demolding of hot embossed polymer microstructures, in: Mas-sachusetts Institute of Technology Department of Mechanical Engineering.Cambridge, MA, 2010.

79] M. Worgull, M. Heckele, New aspects of simulation in hot-embossing, in:Design, Test, Integration and Packaging of MEMS/MOEMS 2003 Symposium,Stresa, Italy, 2003, pp. 272–274.

80] H. Takagi, M. Takahashi, R. Maeda, Y. Onishi, Y. Iriye, T. Iwasaki, et al., Exper-imental and numerical analyses on recovery of polymer deformation after

ators B 190 (2014) 451– 458

demolding in the hot embossing process, Journal of Vacuum Science and Tech-nology B: Microelectronics and Nanometer Structures 26 (2008) 2399–2403.

81] J.-W. Choi, et al., Development and characterization of microfluidic devices andsystems for magnetic bead-based biochemical detection, Biomedical Microde-vices 3 (2001) 191–200.

82] J. Peirs, D. Reynaert, H. van Brussel, Design of miniature parallel manipulatorsfor integration in a self-propelling endoscope, Sensors and Actuators A: Physical85 (2000) 409–417.

83] C.M. Ho, X. Yang, C. Grosjean, Y.C. Tai, A MEMS thermopneumatic siliconerubber membrane valve, Sensors and Actuators A: Physical 64 (1998) 101–108.

84] M. Kohl, D. Dittmann, E. Quandt, B. Winzek, Thin film shape memorymicrovalves with adjustable operation temperature, Sensors and Actuators A:Physical 83 (2000) 214–219.

Biographies

Jaione Etxebarria: She is an M.Sc. in chemical engineering (2008) and a postgradu-ate master’s degree in engineering of advanced materials (2010) from the Universityof the Basque Country in Bilbao, Spain. In 2008 she joined the Microsystems researchgroup at CIC microGUNE/IK4-Ikerlan (Spain) and is currently working toward Ph.D.degree. Her main research interests are focused on design, microfabrication anddevelopment of polymeric microfluidic control devices, microvalves and flow sen-sors for Lab-on-a-chip applications.

Javier Berganzo: He received B.S. degree in electronic engineering from the Mon-dragón University and his MS degree in electric engineering from the Lausanne’sSwiss Federal Institute of Technology (1990). Currently, he is working in theMicrosystems research group at CIC microGUNE/IK4-Ikerlan (Spain). His researchfocuses on microfluidics and system integration for medical devices and disposablediagnosis cartridges.

Jorge Elizalde: He received his M.Sc. and Ph.D. degrees in engineering from the Uni-versity of Navarra (Spain) in 1992 and 1997, respectively. Currently, he is workingin the Microsystems research group at CIC microGUNE/IK4-Ikerlan (Spain). He hadworked on the simulation, design, fabrication and characterization of microfluidicdevices. His research is now focused on the fabrication of the low cost polymericmicrofluidic devices.

Luis José Fernandez: He received the M.Sc. degree in Physics at Zaragoza University(Spain) in 2001. He then joined the Transducers Science and Technology depart-ment at Twente University (The Netherlands) where he obtained the Ph.D. degreein 2005 for the development of MEMS based power sensor for radio frequency sig-nals. During the next 5 years he was involved in the development of low cost MicroTotal Analysis Systems (�TAS), microfluidic control elements and biosensors at CICmicroGUNE/IK4-Ikerlan research institute (Spain), becoming a formal partner of thecentre. In 2010 He received the “Ramon y Cajal Fellowship” to join the GEMM groupat Zaragoza University (Spain) and lead scientific advances in the field of microflu-idic devices for cell culture applications. Dr. Fernandez has published 19 scientificpapers with SCI, 34 articles in proceedings of international conferences, and holds9 patents.

Aitor Ezkerra: He is an M.Sc. in thermal power from Cranfield University

(2003) and Ph.D. in engineering from the University of the Basque Countryin Bilbao, Spain (2009). In 2004 he joined the Microsystems research groupat CIC microGUNE/IK4-Ikerlan (Spain). His main research interests are focusedon microfluidics, microfabrication in polymers, embedded microcantilevers formicrofluidic control, micropumps and microvalves for Lab-on-a-chip applications.