Sensors and Actuators B 246 (2017) 904–909

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical

journal homepage: www.elsevier.com/locate/snb

Fabricating devices with improved adhesion between PDMS and

gold-patterned glass
Jannu Casanova-Moreno a,b,1,2 , Josiah To a,b,1 , Cheng Wei Tony Yang c,d ,
Robin F.B. Turner a,c,e , Dan Bizzotto b,e,∗ , Karen C. Cheung a,b,∗
a
Department of Electrical and Computer Engineering, University of British Columbia, 2332 Main Mall, Vancouver, Canada
b
Advanced Materials and Process Engineering Laboratory, University of British Columbia, 2355 East Mall, Vancouver, Canada
c
Michael Smith Laboratories, University of British Columbia, 2185 East Mall, Vancouver, Canada
d
Department of Chemical and Biological Engineering, University of British Columbia, 2360 Main Mall, Vancouver, Canada
e
Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Gold-patterned glass substrates sealed to a PDMS fluidic layer is an architecture widely employed
Received 25 October 2016 in microfluidic devices. The plasma bonding method is commonly used but presents the disadvan-
Received in revised form 21 January 2017 tage of being ineffective on gold surfaces, creating a sealing weakness where solution leakage can
Accepted 17 February 2017
be a problem. Frequently, alternative approaches to sealing include additional patterned layers that
Available online 24 February 2017
often require expensive cleanroom time. Here we demonstrate a sealing method involving carboxyl-
terminated and amine-terminated silanes appropriate for PDMS-based device fabrication with patterned
Keywords:
gold/glass surfaces. This cleanroom-free methodology improves solution leak pressure 37-fold com-
PDMS–gold bonding
Electrochemical sensor
pared to the standard plasma bonding technique on these patterned substrates. This bonding strategy
Microfluidics is demonstrated in a proof-of-concept electrochemical cell enclosed in a PDMS chamber. An electro-
Surface modification chemical method for preparing a clean and electroanalytically useful gold surface in the PDMS device is
Fluid leakage detailed.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction fluidic spaces (i.e. acting as electrodes), another portion to be

sealed with the polymeric slab, and a final section to be exter-
Electronically conductive materials are frequently integrated nally attached to an electrical connection. Examples of the use
into microfluidic devices for various applications [1]. Electrohy- of several metals can be found in literature, gold being increas-
drodynamics have been used for pumping [2], mixing [3] and ingly employed due to its versatile surface functionalization [11].
separation [4], faradaic reactions for quantification [5,6] and Regarding the fluidic layer materials, although rigid thermoplastics
catalysis [7], electrical resistance for temperature control and mea- are favoured for their industrial production due to the high scalabil-
surement [8], while electrowetting forms the basis of “digital” ity of their manufacturing processes, polydimethylsiloxane (PDMS)
microfluidics [9]. Many of these microfluidic systems are composed is still commonly used in the research setting [12]. PDMS is not only
of patterned metal features on a flat glass or silicon substrate, and cheap and easy to employ but also moldable and flexible, which
a polymeric slab containing chambers and/or channels adhered helps it to conform to non-flat substrates like the metal-patterned
to the hybrid (i.e. patterned) substrate [1,10]. With some excep- ones [1].
tions (e.g. heaters and “digital” microfluidics), most devices require There are a number of reported strategies to bond PDMS slabs
part of the metallic surface to be exposed to the solution in the to glass substrates [13]. The simplest method involves forming a
reversible bond using clamps to maintain the seal [14]. Despite
its apparent simplicity, the use of clamps can be cumbersome if
∗ Corresponding authors. additional equipment is needed in close proximity to the chip, also
E-mail addresses: bizzotto@chem.ubc.ca (D. Bizzotto), kcheung@ece.ubc.ca making transportation less straightforward (e.g. alignment has to
(K.C. Cheung). be maintained). Furthermore, the force exerted by the clamp(s)
1
These authors contributed equally to this work.
2
may deform the PDMS and add internal stress to the device. Treat-
Present address: Laboratorio Nacional de Micro y Nanofluídica, Centro de Inves-
tigación y Desarrollo Tecnológico en Electroquímica, Parque Tecnológico Querétaro
ing the PDMS surfaces (and often the substrates as well) with
s/n, Pedro Escobedo, México. oxygen [15] or air [16] plasma is the most widely used bonding

http://dx.doi.org/10.1016/j.snb.2017.02.109
0925-4005/© 2017 Elsevier B.V. All rights reserved.
 J. Casanova-Moreno et al. / Sensors and Actuators B 246 (2017) 904–909 905

method. These methods involve the creation of hydroxyl groups

on the PDMS surface, permitting the formation of covalent bonds
between the PDMS and glass [15,17]. However, the metal-coated
areas on the substrate will not bond using this method. There-
fore, solution may leak at the PDMS–metal interface which can
affect device performance [18]. A palliative measure is to drive
flow by applying a negative pressure at the chip outlet rather
than a positive one at the inlet [18]. Kohler et al. have reported
the use of a 1-hexadecanethiol self assembled monolayer as an
“adhesion promoter” to bond sputtered Au electrodes to PDMS
using nitrogen plasma [19]. Unfortunately, a detailed description
of the method and its mechanism is not available in their work
or elsewhere. Often an additional layer of a material that can be
plasma-bonded to PDMS (e.g. silicon dioxide [18,20,21] or SU-
8 photoresist [22]) is patterned on top of the hybrid substrate.
This approach, however, involves additional cleanroom time, spe-
cialized equipment and/or complicated etching procedures. Other
methods make use of an adhesive layer [23–25], where the main
challenges are obtaining a uniform thickness and avoiding clogging
of the channels during bonding [13]. On the other hand, mercap-
tosilanes have been used to bond PDMS to metals (e.g. gold or silver)
[26–29], but they are unable to bond PDMS to glass surfaces mak-
Fig. 1. Schematic of a PDMS-based 3-electrode device. (a) Top view of the PDMS-
ing this approach not suitable for hybrid or patterned substrates. bonded Au/glass hybrid chip device. (b) Cross-sectional view of the hybrid device
For these microfluidic systems that integrate metallic patterns on through the center (i.e., down the length) of the device. (c) Enlarged view of the
glass, developing a simple and easy fabrication method that leads proposed carboxyl-amine chemistry between PDMS and the gold/glass substrate.
to stable, leak-free, and reliable devices still remains challenging CE, RE and WE stand for counter, reference and working electrodes, respectively.

[30].
Previously, the combined use of the carboxylated silane 2. Material and methods
N-[(3-trimethoxysilyl)propyl]-ethylene-diamine triacetic acid
(TMS-EDTA) and the amminated silane (3-aminopropyl)- The 3-electrode substrates were fabricated using standard lift-
trimethoxysilane (3-APTMS) has been shown to facilitate the off techniques. A 10 nm Ti layer was used to improve the adhesion
bonding between PDMS and either a glass or gold surface [31]. By of the 200 nm thick Au features onto a Borofloat 33 glass wafer
chemically modifying a PDMS surface with 3-APTMS and a glass (10 cm diameter and 500 ␮m thickness, University Wafer; Boston,
or gold surface with TMS-EDTA, carboxyl-amine bonds can form MA). The glass wafers were then diced into smaller chips (Fig. 1a)
between the PDMS (functionalized with terminal primary amino designed to fit inside a DropSens ␮STAT Cable Connector. The Au
groups) and the glass or gold slide (functionalized with terminal electrode leads had a width of 770 ␮m or 500 ␮m. In order to cre-
carboxyl groups). While silanes are known to chemisorb on sub- ate a sufficiently smooth, flat bonding surface between the PDMS
strates containing terminal OH groups [32] (such as UV-ozone microfluidic chamber and the glass/gold substrate, a 2-step cast-
treated glass and PDMS [33]), the adsorption of TMS-EDTA on gold ing process was used (Figs. S1 and S2). After casting, each PDMS
is not obvious and required further study. Characterization by chamber was hole-punched to create inlet and outlet ports.
several surface chemical analysis techniques and electrochemical
methods revealed measured free energies of adsorption of ∼−20 to 2.1. Substrate modification
−30 kJ/mol in a complex buffer [34]. Furthermore, it was observed
that the interaction is potential-dependent, with minimal adsorp- The glass, gold-coated glass and patterned 3-electrode sub-
tion at very negative potentials (∼−1.1 V vs a saturated calomel strates were first cleaned by sequential sonication in acetone,
electrode). These results suggest that TMS-EDTA can be used to 2-propanol, methanol and ultrapure water (15 min each). The sub-
bond PDMS with both glass and gold surfaces. The remaining strates were then immersed for 15 min in RCA1 solution (5:1:1
carboxylic groups on the gold surface can be used for further volume of ultrapure water, NH4 OH, and H2 O2 ) at room temperature
functionalization, or be electrochemically removed to recover the followed by immersion in 0.1 M HNO3 for 30 min [35]. The cleaned
bare Au surface. substrates were rinsed with ultrapure water and blown dry in an
Although the as-modified electrodes with the intact adsorbed argon stream. Cleaned substrates were exposed to UV-ozone (UVO-
TMS-EDTA can be used as heaters or as electrical interconnects, Cleaner, model 42, Jelight Co. Inc.; Irvine, CA) for 5 min [31]. They
where their surface characteristics may not significantly affect were then immersed in an aqueous solution of 10% TMS-EDTA (v/v)
their performance, their use as electrodes in an electrochemical for 2 h at room temperature. TMS-EDTA-modified substrates were
cell requires further method development and characterization. washed three times in ultrapure water and dried under a nitrogen
To demonstrate and evaluate the applicability of the described or argon stream.
methodology in device fabrication, this work presents the con- The PDMS surface was cleaned using 3M Scotch Magic Tape and
struction of a three electrode electrochemical cell. The basis of this exposed to UV–Ozone for 5 min. They were immediately immersed
device is a borosilicate glass substrate patterned with Au electrodes into a 10% (v/v) solution of 3-APTMS in ultrapure water for 1 h. The
(and connectors) irreversibly bonded to a PDMS chamber through PDMS surface was then washed three times with water and dried
the TMS-EDTA/3-APTMS interaction (Fig. 1). The strength of this for immediate use.
carboxyl-amine bonding strategy is compared with the air plasma Substrates modified with TMS-EDTA were acti-
bonding method by conducting fluid pressure leak tests on glass, vated using carbodiimide chemistry. Aqueous solutions
gold and patterned surfaces. Finally, different methods to electro- of 0.1 M N-hydroxysuccinimide (NHS) and 0.4 M N-(3-
chemically clean TMS-EDTA from the Au surface inside a bonded Dimethylaminopropyl)-N -ethylcarbodiimide hydrochloride
device were tested. (EDC) were prepared separately, then mixed 1:1 v/v immediately
906 J. Casanova-Moreno et al. / Sensors and Actuators B 246 (2017) 904–909

prior to use. The solid substrates were immersed in this mixture Table 1
Comparison of leak pressures using different bonding strategies. At least five repli-
for 30 min to activate the terminal carboxyl groups, followed by
cates were used to calculate the mean and standard deviation.
rinsing with ultrapure water and drying with N2 or Ar. The acti-
vated substrates were immediately brought into contact with the Bonding strategy Glass Gold Au/glass
3-APTMS-modified PDMS surfaces, and allowed to react at room Air plasma >180 kPa On loading 1 ± 1 kPa
temperature for 16 h. For control experiments, the surfaces of both Carboxyl-amine 134 ± 16 kPa 31 ± 13 kPa 37 ± 10 kPa
unmodified PDMS and clean solid substrates (glass, gold-coated
glass or 3-electrode substrate) were exposed to air plasma for
1.25 min, then brought into contact and let sit at room temperature
the measurement of the leak pressure (i.e. the maximum pressure
for 16 h.
a bonded chamber can withstand) to determine if the device can
be stable during solution filling or exchange. Fluid pressure leak
2.2. Fluid leak tests
tests were performed on PDMS chambers bonded to glass, gold,
and patterned hybrid Au/glass substrates. An example of the non-
The strength of the PDMS–substrate interaction was evaluated
linear pressure increase during this process is shown in ESI (Fig.
by performing fluid pressure leak tests using the experimental
S4) along with observations of the initial stages of fluid leakages
setup shown in the ESI (Fig. S3). A 2 megapixel digital colour
at each of the three PDMS/Au interfaces. Table 1 shows the results
camera (A3250U, OMAX) with a 0.5X reduction (A3RDF50, OMAX)
from pressure leak tests for the carboxyl-amine and the air plasma
lens was mounted on a stereo zoom microscope (SZ6045, Olym-
bonding strategies.
pus). OMAX ToupView 3.7 software was used to record images and
The PDMS–glass devices which were air plasma-bonded showed
videos of the pressure leak tests. Open-source hardware, Arduino
a leak pressure that surpassed the pressure sensor detection limit of
ATmega328, was used to process the pressure sensor informa-
180 kPa. This is consistent with literature values of ∼200–350 kPa
tion and send the data to a computer. Each PDMS chamber was
[37,38]. On the other hand, PDMS chambers treated with air plasma
manually loaded by injection of green dye solution at low pres-
did not bond with similarly treated bare Au substrates as the solu-
sure. After purging the chamber, the outlet port was blocked
tion started to leak upon loading of the green dye. These results
while the inlet was connected, via a 3-way valve, to a syringe
show that there is no strong interaction between PDMS and Au.
pump (KDS230, KD Scientific; Holliston, MA). The chamber was
Leaks in hybrid Au/glass substrates occurred near the loading pres-
placed on the microscope stage, and the syringe pump was used
sure at 1 ± 1 kPa. Even though the air plasma bonding method
to drive a continuous 1 mL/min flow, which in turn increased
yielded a strong PDMS–glass bond, the solution leaked through the
the pressure in the solution-filled PDMS chamber. The pres-
PDMS–gold interfaces (Fig. 2a–c).
sure was monitored by the pressure sensor (HDIB002GUSMD8P5,
The seal between glass slides and PDMS chambers bonded
First Sensor/Sensortechnics; Mansfield, MA) until the maximum
using the carboxyl-amine strategy was more resilient than the
measurable pressure was achieved. For samples that leaked, the
one between gold and PDMS using the same approach. This is in
pressure was recorded at the first sign of leakage (e.g., when the
line with the fact that TMS-EDTA is physically adsorbed onto Au
green dye emerges at either the PDMS–glass or PDMS–Au inter-
from a complex electrolyte [34] while the siloxane bonds formed
face) as determined from the video images. For each bonding
with glass are a stronger chemisorption interaction (∼450 kJ/mol)
method/substrate combination, Grubbs’ test was performed on six
[39]. Using the carboxyl-amine bonding strategy, PDMS–Au and
replicate leak pressure measurements to identify (and remove) out-
PDMS–Au/glass interfaces presented comparable leak pressures,
liers [36].
confirmed by the observation that, in the latter, the solution first
leaked at the PDMS–gold interface. Sample images from these leak
2.3. Electrochemical cleaning
tests are shown in Fig. 2d–f. These results show that the carboxyl-
amine-bonded devices using patterned Au/glass substrates were
To remove the surface modification from the gold electrode,
able to remain bonded at fluid pressures that were about 37×
electrochemical desorption was performed on bonded devices
higher than the substrates that used the air plasma strategy, due
to prepare a bare Au surface. In this proof-of concept experi-
to the increased interaction between PDMS and gold. Although
ment a chloridized Ag wire connected through the outlet port
this bonding strategy is not suitable for the full pressure range
was used as the reference electrode; alternatively Ag can be
used in some applications of microfluidic devices [40], a significant
deposited on one of the patterned electrodes to keep the entire
number of applications do not need to withstand these high fluid
system contained. The reduction of 0.75 mM hexamminecobalt(III)
pressures.
(Co(NH3 )6 3+ , or CoHex) (in a 100 mM phosphate buffer electrolyte)
was used to assess the surface cleanliness for electrochemical
faradaic reactions. This required removing oxygen which was done 3.2. Electrochemical testing
by enclosing the entire chip inside a container deoxygenated
with argon. The details of the experimental setup are given in The utility of the carboxyl-amine bonding strategy for gold
the ESI (Fig. S5). For comparison, CoHex reduction was mea- patterned onto glass is demonstrated through the preparation and
sured on a clean gold bead electrode in a conventional 3-electrode testing of an electrochemical cell enclosed by a PDMS chamber.
cell. After bonding the device using the carboxyl-amine strategy, the Au
electrodes are still modified with TMS-EDTA (Fig. 1c), which may
3. Results and discussion negatively influence electrochemical detection since the adsorbed
layer would impede the electron transfer between electroactive
3.1. Fluid leak tests species in solution and the electrode surface. Thus, it is important to
devise a method for cleaning the gold surface by removing adsorbed
Shear tests conducted previously using the TMS-EDTA/3-APTMS TMS-EDTA in the assembled device. Cyclic voltammograms (CV) of
system on separate glass and gold substrates [31] suggested the an aqueous CoHex solution were used as an indication of the clean-
possibility of bonding hybrid Au/glass substrates to PDMS using liness of the gold electrode. For comparison, a gold bead electrode
the same carboxyl-amine chemistry. However, a realistic metric was cleaned via flame annealing, a procedure generally accepted to
of the bond strength for its use in microfluidic devices requires generate clean electrode surfaces [41]. In these conditions CoHex
 J. Casanova-Moreno et al. / Sensors and Actuators B 246 (2017) 904–909 907

Fig. 2. Images of PDMS chambers bonded to hybrid Au/glass substrates during leak tests at selected fluid pressures. Top row: (a) air plasma bonded device at atmospheric
pressure after loading of the green dye. (b) Leakage of plasma-bonded device at PDMS–gold interface. (c) Complete failure of plasma bonded device. Bottom row: device
bonded with carboxyl-amine strategy at (d) atmospheric pressure after dye loading, (e) 3 kPa (i.e. above the leakage pressure of plasma-bonded device), and (f) 34 kPa (initial
solution leakage). The dotted outline highlights the area in which the leakage is observed.

vs Ag|AgCl wire, each followed by 10 s at −0.35 V) in 0.1 M KOH.

After pulsing, the reductive peak potential for CoHex shifted pos-
itively to ∼−0.13 V (red dashed curve) indicating a decrease in
adsorbed TMS-EDTA impeding the reduction. Further treatment by
performing CVs (between −0.16 and 1.47 V vs Ag|AgCl wire) in 1 M
H2 SO4 (green dotted line), a widespread electrochemical cleaning
method, yields a surface with a corresponding CoHex reduction
peak at ∼−24 mV, similar to the flame annealed and cleaned bead
electrode.
These measurements demonstrate that it is possible to electro-
chemically clean the patterned electrodes in a bonded device so as
to produce a clean gold surface. This cleaning approach does result
in a surface that shows a more rapid decrease in CoHex reduction
current with time when compared to the same measurement using
a gold bead electrode (Fig. S6), presumably due to readsorption of
the contaminants from the much smaller solution volume in the
device, in addition to other differences which will be subject to
Fig. 3. Cyclic voltammograms of 750 ␮M hexamminecobalt(III) solutions on gold further study.
electrodes. Working electrodes were a fabricated device at different stages of the
cleaning procedure (as annotated on the legend) and a flamed gold bead. Poten-
tial scan rate = 20 mV/s. First cycle is shown. The electrolyte was a deoxygenated
100 mM phosphate buffer solution without added chloride. Regardless of the ref- 4. Conclusions
erence electrode used for a particular experiment, the potentials are reported with
respect to Ag|AgCl|NaCl 3 M. (For interpretation of the references to colour in the
citation of this figure, the reader is referred to the web version of the article.)
We have shown that TMS-EDTA can be used as a modifying
agent for “hybrid” or gold-patterned glass surfaces. These modi-
fied surfaces react readily with 3-APTMS-modified PDMS slabs to
presents a highly irreversible reduction reaction with a peak poten- form sealed devices which include electrodes and electrical con-
tial (Ep ) at −24 mV vs Ag|AgCl|NaCl 3 M (Fig. 3). On partially blocked tact leads. When compared with the common plasma bonding
electrodes, reduction reactions of this type present a large Ep shift technique, the presented method produced devices that resisted
to more negative potentials [42]. leakage until higher fluid pressures. This improvement can be
Fig. 3 shows CVs recorded on the gold electrode in a bonded explained through an increase in the strength of the interactions
device before and after several cleaning procedures. The CV for between gold and PDMS surfaces when modified with TMS-EDTA
an as-fabricated device (blue dashed line) shows a CoHex reduc- and 3-APTMS respectively, as compared with the air plasma-
tion wave at ∼−0.3 V which is significantly shifted in the negative treated unmodified Au and PDMS. Devices containing both gold and
direction with respect to that of the clean bead (solid line). glass surfaces bonded using the carboxyl-amine strategy withstood
Previous differential capacitance studies show that TMS-EDTA a pressure 37 times higher than the plasma-bonded devices, both
only weakly adsorbs at very negative potentials [34] and can be devices leaking at the Au–PDMS interface. In addition it was shown
removed at potentials more negative than −0.9 V vs SCE [43] that electrochemical cleaning adequately removed TMS-EDTA from
under ideal conditions. Electrochemical desorption of TMS-EDTA the parts of the Au pattern not in contact with the PDMS slab. There-
was attempted by applying 200 potential pulses (0.5 s @ −1.6 V fore we propose this methodology as a cleanroom-free alternative
908 J. Casanova-Moreno et al. / Sensors and Actuators B 246 (2017) 904–909

to improve the adhesion of PDMS fluidic layers to Au-patterned [26] M. Kim, S.-Y. Lee, H. Choi, Y.-B. Shin, S.O. Jung, M.-G. Kim, B.H. Chung, On-chip
glass supports. Escherichia coli culture, purification, and detection of expressed proteins, Eur.
Biophys. J. 35 (8) (2006) 655–662.
[27] K.J. Lee, K.A. Fosser, R.G. Nuzzo, Fabrication of stable metallic patterns
Acknowledgements embedded in poly(dimethylsiloxane) and model applications in non-planar
electronic and lab-on-a-chip device patterning, Adv. Funct. Mater. 15 (4)
(2005) 557–566.
This work was supported by a grant from NSERC (Canada) and [28] K.S. Lim, W.-J. Chang, Y.-M. Koo, R. Bashir, Reliable fabrication method of
CIHR (Canada) through the Collaborative Health Research Projects transferable micron scale metal pattern for poly(dimethylsiloxane)
metallization, Lab Chip 6 (2006) 578–580.
Programme. [29] S. Schuy, A. Janshoff, Microstructuring of phospholipid bilayers on gold
surfaces by micromolding in capillaries, J. Colloid Interface Sci. 295 (1) (2006)
93–99.
Appendix A. Supplementary Data [30] A. TermehYousefi, S. Bagheri, N. Adib, Integration of biosensors based on
microfluidic: a review, Sens. Rev. 35 (2) (2015) 190–199.
Supplementary data associated with this article can be found, in [31] E. Ouellet, C.W.T. Yang, T. Lin, L.L. Yang, E.T. Lagally, Novel carboxyl-amine
bonding methods for poly(dimethylsiloxane)-based devices, Langmuir 26
the online version, at http://dx.doi.org/10.1016/j.snb.2017.02.109. (14) (2010) 11609–11614.
[32] G.T. Hermanson, Chapter 13 - Silane Coupling Agents, in: G.T. Hermanson
(Ed.), Bioconjugate Techniques (third edition), Academic Press, Boston, 2013,
References pp. 535–548.
[33] Y.-J. Fu, H.-z. Qui, K.-S. Liao, S.J. Lue, C.-C. Hu, K.-R. Lee, J.-Y. Lai, Effect of
[1] E. Verpoorte, N.F. De Rooij, Microfluidics meets MEMS, Proc. IEEE 91 (6) UV-ozone treatment on poly(dimethylsiloxane) membranes: surface
(2003) 930–953. characterization and gas separation performance, Langmuir 26 (6) (2010)
[2] A. Ramos, Microfluidic Technologies for Miniaturized Analysis Systems, 4392–4399.
Springer US, Boston, MA, 2007, pp. 59–116. [34] C.W.T. Yang, I. Martens, E.L. Gyenge, R.F.B. Turner, D. Bizzotto, Adsorption of a
[3] N.-T. Nguyen, Microfluidic Technologies for Miniaturized Analysis Systems, carboxylated silane on gold: characterization for its rational use in hybrid
Springer US, Boston, MA, 2007, pp. 117–155. glass/gold substrates, J. Phys. Chem. C 120 (5) (2016) 2675–2683.
[4] V.M. Ugaz, J.L. Christensen, Microfluidic Technologies for Miniaturized [35] D. Hohertz, S.F. Romanuik, B.L. Gray, K.L. Kavanagh, Recycling gold nanohole
Analysis Systems, Springer US, Boston, MA, 2007, pp. 393–438. arrays, J. Vac. Sci. Technol. A 32 (3) (2014) 031403.
[5] P.B. Lillehoj, M.-C. Huang, N. Truong, C.-M. Ho, Rapid electrochemical [36] F.E. Grubbs, Procedures for detecting outlying observations in samples,
detection on a mobile phone, Lab Chip 13 (2013) 2950–2955. Technometrics 11 (1) (1969) 1–21.
[6] D.G. Rackus, M.H. Shamsi, A.R. Wheeler, Electrochemistry, biosensors and [37] J.C. McDonald, D.C. Duffy, J.R. Anderson, D.T. Chiu, H. Wu, O.J.A. Schueller,
microfluidics: a convergence of fields, Chem. Soc. Rev. 44 (2015) 5320–5340. G.M. Whitesides, Fabrication of microfluidic systems in
[7] E. Kjeang, N. Djilali, D. Sinton, Microfluidic fuel cells: a review, J. Power poly(dimethylsiloxane), Electrophoresis 21 (1) (2000) 27–40.
Sources 186 (2) (2009) 353–369. [38] S. Bhattacharya, A. Datta, J.M. Berg, S. Gangopadhyay, Studies on surface
[8] V. Miralles, A. Huerre, F. Malloggi, M.-C. Jullien, A review of heating and wettability of poly(dimethyl) siloxane (PDMS) and glass under
temperature control in microfluidic systems: techniques and applications, oxygen-plasma treatment and correlation with bond strength, J.
Diagnostics 3 (1) (2013) 33. Microelectromech. Syst. 14 (3) (2005) 590–597.
[9] H. Moon, C.-J.C. Kim, Microfluidic Technologies for Miniaturized Analysis [39] J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry: Principles of
Systems, Springer US, Boston, MA, 2007, pp. 203–239. Structure and Reactivity, HarperCollins College Publishers, New York, NY,
[10] M. Pumera, A. Merkoçi, S. Alegret, New materials for electrochemical sensing 1993.
VII. Microfluidic chip platforms, Trends Anal. Chem. 25 (3) (2006) [40] N.T. Nguyen, S.T. Wereley, Fundamentals and Applications of Microfluidics,
219–235. Artech House Integrated Microsystems Series, Artech House, 2006.
[11] E. Prats-Alfonso, F. Albericio, Functionalization of gold surfaces: recent [41] L.A. Kibler, Preparation and Characterization of Noble Metal Single Crystal
developments and applications, J. Mater. Sci. 46 (24) (2011) 7643–7648. Electrode Surfaces, International Society of Electrochemistry, 2003.
[12] E.K. Sackmann, A.L. Fulton, D.J. Beebe, The present and future role of [42] C. Amatore, J.M. Savéant, D. Tessier, Charge transfer at partially blocked
microfluidics in biomedical research, Nature 507 (7491) (2014) 181–189. surfaces: a model for the case of microscopic active and inactive sites, J.
[13] Y. Temiz, R.D. Lovchik, G.V. Kaigala, E. Delamarche, Lab-on-a-chip devices: Electroanal. Chem. 147 (1-2) (1983) 1–38.
how to close and plug the lab? Microelectron. Eng. 132 (2015) 156–175. [43] Z.L. Yu, C.W.T. Yang, E. Triffaux, T. Doneux, R.F.B. Turner, D. Bizzotto,
[14] T. Robinson, P.E. Verboket, K. Eyer, P.S. Dittrich, Controllable electrofusion of Measuring and remediating nonspecific modifications of gold surfaces using a
lipid vesicles: initiation and analysis of reactions within biomimetic coupled in situ electrochemical fluorescence microscopic methodology, Anal.
containers, Lab Chip 14 (2014) 2852–2859. Chem. 89 (1) (2016) 886–894.
[15] D.C. Duffy, J.C. McDonald, O.J.A. Schueller, G.M. Whitesides, Rapid prototyping
of microfluidic systems in poly(dimethylsiloxane), Anal. Chem. 70 (23) (1998)
4974–4984.
[16] J.C. McDonald, G.M. Whitesides, Poly(dimethylsiloxane) as a material for
fabricating microfluidic devices, Acc. Chem. Res. 35 (7) (2002) 491–499.
[17] E. Wilhelm, C. Neumann, K. Sachsenheimer, K. Länge, B.E. Rapp, Rapid
bonding of polydimethylsiloxane (PDMS) to various stereolithographically Biographies
(STL) structurable epoxy resins using photochemically cross-linked
intermediary siloxane layers, in: B.L. Gray, H. Becker (Eds.), SPIE
MOEMS-MEMS, SPIE, 2014, 89760A-89760A-7. Jannu Casanova-Moreno received his B.Sc. (2006) and Ph.D. (2014) degrees from
[18] N.H. Moreira, A.L. de Jesus de Almeida, M.H. de Oliveira Piazzeta, D.P. de Jesus, the School of Chemistry of the National Autonomous University of Mexico and the
A. Deblire, A.L. Gobbi, J.A. Fracassi da Silva, Fabrication of a multichannel Chemistry Department of the University of British Columbia, respectively. He is now
PDMS/glass analytical microsystem with integrated electrodes for working as a professor at the Center for Research and Technological Development
amperometric detection, Lab Chip 9 (2009) 115–121. in Electrochemistry (CIDETEQ).
[19] S. Kohler, C. Weilbeer, S. Howitz, H. Becker, V. Beushausen, D. Belder, PDMS
Josiah To received his B.S. degree in bioengineering from University of California
free-flow electrophoresis chips with integrated partitioning bars for bubble
San Diego (UCSD) in La Jolla, California in 2013 and completed a MASc in biomedical
segregation, Lab Chip 11 (2011) 309–314.
engineering in 2016 at the University of British Columbia. He is currently a medical
[20] K.C. Tang, E. Liao, W.L. Ong, J.D.S. Wong, A. Agarwal, R. Nagarajan, L. Yobas,
student at the Chicago Medical School at Rosalind Franklin University of Medicine
Evaluation of bonding between oxygen plasma treated polydimethyl siloxane
and Science.
and passivated silicon, J. Phys.: Conf. Ser. 34 (1) (2006) 155.
[21] X. Li, K.G. Klemic, M.A. Reed, F.J. Sigworth, Microfluidic system for planar Cheng Wei Tony Yang has received his B.Sc. degree (2008) in Physics and Life Sci-
patch clamp electrode arrays, Nano Lett. 6 (4) (2006) 815–819. ences and Ph.D. degree (2016) in Chemical and Biological Engineering from The
[22] Z. Zhang, P. Zhao, G. Xiao, B.R. Watts, C. Xu, Sealing SU-8 microfluidic channels University of British Columbia. He is currently the Program Manager of Providence
using PDMS, Biomicrofluidics 5 (4) (2011) 046503. Airway Centre at HLI, St Paul’s Hospital.
[23] H. Wu, B. Huang, R.N. Zare, Construction of microfluidic chips using
polydimethylsiloxane for adhesive bonding, Lab Chip 5 (2005) 1393–1398. Robin F.B. Turner earned a Ph.D. degree in electrical engineering from the Univer-
[24] S. Satyanarayana, R.N. Karnik, A. Majumdar, Stamp-and-stick sity of Alberta, Edmonton, AB, Canada, in 1990. He is a Professor at The University
room-temperature bonding technique for microdevices, J. Microelectromech. of British Columbia with joint appointments in the Michael Smith Laboratories and
Syst. 14 (2) (2005) 392–399. Department of Electrical & Computer Engineering; he is also an Associate Member
[25] M. Díaz-González, A. Baldi, Fabrication of biofunctionalized microfluidic of the Department of Chemistry. His main research activities focus on applica-
structures by low-temperature wax bonding, Anal. Chem. 84 (18) (2012) tions of molecular (vibrational) spectroscopy to analytical problems in biochemistry,
7838–7844. biotechnology, and biomedical engineering.
 J. Casanova-Moreno et al. / Sensors and Actuators B 246 (2017) 904–909 909

Dan Bizzotto received an BSc(Hon) in Applied Chemistry from the Univer- Karen C. Cheung received the B.S. and Ph.D. degrees in Bioengineering from
sity of Calgaryand a PhD degree from the University of Guelph followed by the University of California, Berkeley, USA. She was a Postdoctoral Researcher in
postdoctoral stays in Plymouth UK and at the FHI (Berlin). He is currently a pro- Microtechnology at the ÉcolePolytechnique Fédérale de Lausanne, Switzerland. She
fessor of Chemistry at the University of British Columbia. His research interests is now at the University of British Columbia. Her research interests include lab-
include study of modified electrode surface using in-situ spectroelectrochemical on-a-chip systems for cell culture and characterization, inkjet printing for tissue
methods. engineering, and implantable neural interfaces.