Microfluid Nanofluid (2009) 7:819–827
DOI 10.1007/s10404-009-0440-7
RESEARCH PAPER
A regenerative biosensing surface in microfluidics
using electrochemical desorption of short-chain
self-assembled monolayer
Seokheun Choi Æ Junseok Chae
Received: 20 February 2009 / Accepted: 26 March 2009 / Published online: 10 April 2009
Ó Springer-Verlag 2009
Abstract We report a novel method to regenerate a
biosensor surface in microfluidics. By applying a low DC
voltage (0.9 V) between two electrodes submerged in
phosphate buffered saline, the sensing surface resets to be
reusable and reconfigurable; streptavidin-bound COOH–
SAM completely desorbs and CH3-terminated self-assem-
bled monolayer (SAM) forms on the sensing surface to
capture the subsequent target molecule, fibrinogen, in a
microfluidic device. The biomolecular interactions are
monitored by surface plasmon resonance in real time, and
ellipsometry and linear sweep voltammetry are used to
evaluate the results. Despite much study on the theoretical
mechanism of electrochemical SAM desorption, relatively
little research has been carried out on its full integration
into a microfluidic system. This is because of electrode
peeling-off and electrolysis occurring at a similar potential
to the potential of the SAM desorption. In this paper, we
report that the potential for the reductive desorption of thiol
SAMs depends on the length of the alkyl chin, the type of
terminal groups and the binding of proteins and that our
approach using short-chain SAMs (n \ 3) can be a good
candidate to minimize these limitations. While the surface
modified by proteins on the long-chain SAMs (n [ 10)
needs more than 1.0 V between two electrodes to be
completely regenerated, the protein-bound surface on the
short-chain SAMs (n \ 3) does around 0.9-DC voltage.
Linear sweep voltammetry demonstrates that hydrogen
evolution (approx. -1.2 V) does not overlap the short-
chain SAM desorption and the electrode does not peel off
during the desorption process as well. It is shown that the
S. Choi (&)
J. Chae
Department of Electrical Engineering, Arizona State University,
Tempe, AZ, USA
e-mail: shchoi2@asu.edu
modified proteins in the same microfluidic device are still
stable even after 10 cycles showing a relative standard
deviation lower than 1.86%.
Keywords Self-assembled monolayer

Electrochemical reductive desorption
Biosensor

Regenerative
Surface plasmon resonance
1 Introduction
Biosensors have increasingly become relevant in their
application to areas as diverse as medicine, biotechnology,
environmental monitoring, food industry, and even mili-
tary technology. Recently, there have been many efforts to
transform the conventional biosensors into a lab-on-a-chip
by introducing microfluidics. This is because microfluidics
provide benefits including low consumption of costly
reagents, minimal handling of hazardous materials, short
reaction time required for analysis, multiple sample
detection in parallel, portability, and versatility in design
(Barry et al. 2004). Moreover, microfluidics technologies
have received much more attention in the biosensor
communities because of their disposable nature. However,
to the quantitative applications using microfluidics, rela-
tively few studies have been devoted. Since quantitative
technique requires detailed calibration of biosensors,
quantitation has been typically obtained using a series of
measurements with disposable sensor units (Asanov et al.
1998), nonetheless variation coming from batch-to-batch
irreproducibility and uncontrollable experimental condi-
tion generates unalterable false-positive/negative respon-
ses. This motivates this work to design a reusable/
regenerative or even reconfigurable biosensor, especially
in a microfluidic device.
123
820
One technique of implementing reusable sensors is to
remove the linker molecules with all bounded biomole-
cules. Generally, linker molecules are used to immobilize
the bioreceptors on a solid surface to maintain permanent
bonding. Self-assembled monolayer is a good candidate as
a linker molecule because of the ease with which it can be
formed spontaneously and used to control the orientation of
the bioreceptors. To remove the linker molecules, existing
techniques use acid, base or detergent (Deval et al. 2004;
Limbut et al. 2006; Minunni et al. 2004; Gau et al. 2001;
Zhang et al. 2008). However, the extreme solutions damage
the sensing sites and have a fatal impact on the next
regeneration step. Recently, an alternative method has been
proposed using electrochemical desorption of SAMs
(Balasubramanian et al. 2006; Canaria et al. 2006; Mirsky
2002). Its theoretical mechanism has long been studied and
well established (Han et al. 2008; Imabayashi et al. 1997;
Lee et al. 2007; Oyamatsu et al. 2008; Vericat et al. 2001;
Widrig et al. 1991). Electrochemical desorption of SAMs
on metal surfaces occurs at negative potential or positive
potential. At negative potential, a reductive desorption on a
metal (Au) surface occurs according to the following
reaction.
RSAu þ e ! RS þ Au, ð1Þ
where R–S represent an n-alkanethiols, i.e., alkanethiolates
are desorbed from a gold surface by the one-electron
reduction process in an electrolyte. However, no study has
yet demonstrated that SAM desorption can be practically
useful as reusable and reconfigurable biosensors, especially
in microfluidic systems. This is because (1) when reductive
potential is applied to the SAM-coated electrode, the
opposite electrode suffers from corrosion (Santini et al.
2000; Lukaszewski et al. 2006) and (2) hydrogen evolution
reaction (HER) occurs at a similar potential to the potential
of the SAM desorption (Schilardi et al. 2006). It is difficult
to remove hydrogen bubbles by low flow rate of solution;
typically several 10’s of ll/min is used in microfluidic
devices. The bubbles reduce the sensitivity, reliability, and
accuracy of the biosensor. Furthermore, they might affect
the structure of the regenerated SAMs of next cycle. Sev-
eral groups desorbed SAM in a strong alkaline electrolyte
to avoid HER (Lee et al. 2007); yet, most biosensors need
to operate in a physiologically relevant medium such as
PBS (pH = 7.4) and few research activities exist to explore
SAM desorption in such physiological solutions.
In this paper, we present a technique to form reusable/
reconfigurable microfluidic biosensors in PBS. Metal (Au)
electrode corrosion and HER are attributed to high reduc-
tive potential for electrochemical desorption of SAMs,
especially in a physiological solution. In order to minimize
these limitations, we shift the reduction potential less than
-1.0 V (to the more postive direction, such as -0.9 V)
123
Microfluid Nanofluid (2009) 7:819–827
using short-chain SAMs (COOH(CH2)2SH and
CH3(CH2)3SH). SAMs desorb in sharp voltammetric peaks
whose peak potentials shift in the negative potential
direction as the hydrocarbon chain length of the alkanethiol
molecule increases; thus the shorter the chain length is, the
lower the reduction potential is (Imabayashi et al. 1997).
Short-chain SAMs are often used as linker molecules to
immobilize molecular probes including antibodies; thus
our approach can be used for many existing biosensors
(Akram et al. 2004; Wang et al. 2008; Pyun et al. 2005).
Linear sweep voltammetry was used to characterize the
desorption of SAMs as a function of the length of the alkyl
chain, the type of terminal groups and the binding of
proteins. Ellipsometry was used to evaluate the desorption
of SAMs before and after the desorption. SPR in situ
monitored successive experimental procedures of recon-
figurable biosensor in real time, including SAM formation,
protein modification, and SAM desorption in the micro-
fluidic device. In Sect. 2, experimental methods and device
fabrication are discussed. All measurements results and
discussions are presented in Sect. 3. Finally, concluding
remarks follow in Sect. 4.
2 Experimental
2.1 Chemicals
n-Alkanethiols (CH3(CH2)nSH) with n = 3 (1-butanethiol)
and n = 11 (1-dodecanethiol) and x-carboxylic acid al-
kanethiols (COOH(CH2)nSH) with n = 2 (3-mercaptopro-
pionic acid) and n = 10 (11-mercaptoundecanoic acid)
were purchased from Sigma–Aldrich. PBS 19. (1.15 g/L
Na2HPO4, 0.20 g/L KCl, 0.20 g/L KH2PO4, 8.0 g/L NaCl,
pH 7.4), sodium acetate (NaAc), absolute elthanol were
obtained from Mediatech, Inc. 1-ethyl-3-(3-dimethylami-
nopropyl) carbodiimide (EDC) and N-hydroxysuccinimide
(NHS) were purchased from Fisher Scientific. Streptavidin
and fibrinogen were from Calbiochem. All the chemicals
were used without further purification.
2.2 Surface preparation for ellipsometry and linear
sweep voltammetry
Glass substrates (BK7, n = 1.517) were first cleaned in
piranha solution (a 3:1 ration of H2SO4 and H2O2) for
10 min. The cleaned glass was coated with Cr/Au (2 nm/
47 nm) by thermal evaporation. The substrates were
cleaned by oxygen plasma (Harrick Plasma Inc.) at 18 W
for 1 min and then immersed in an ethanol solution of
alkanethiols at 2 mM for 1 h at room temperature to
form SAMs. On COOH-terminated SAMs, streptavidin
was covalently immobilized using EDC/NHS mixture.
Microfluid Nanofluid (2009) 7:819–827
On CH3-terminated SAMs, fibrinogen was adsorbed
via hydrophobic attraction. Each step was described at
Sect. 2.4.2.
2.3 Microfluidic device fabrication
Figure 1 shows the microfluidic device to characterize a
reusable and reconfigurable biosensor. Top and bottom
glass substrates had Cr/Au electrodes and the spacer
between the two substrates was made of polydimethyl
siloxane (PDMS). These layers were bonded all together by
using oxygen plasma. Cr/Au of bottom layer (2 nm/47 nm)
was evaporated and patterned on a 150-lm-thick glass
substrate for SPR measurement. 1 mm-thick top glass was
mechanically drilled to have six holes including two inlets,
two outlets, and two electrical contacts which were filled
with silver paste for feedthroughs. The PDMS layer was
prepared to be 1 mm in thickness and mechanically cut to
have two channels: reference and sample channels. Elec-
trical potential was applied between the top and bottom
electrodes; negative potential to the bottom with respect to
the top electrode.
2.4 Methods and evaluation
The electrochemical desorption process of SAMs has been
investigated by many techniques including electrochemis-
try, Fourier transform infrared spectroscopy (FTIR), elec-
trochemical quartz crystal microbalance (QCM), in situ
scanning tunneling microscope (STM), second harmonic
generation spectroscopy (SHG), sum frequency generation
(SFG), ellipsometry, Raman spectroscopy, and SPR (Ima-
bayashi et al. 1997; Yang et al. 1997; Qu et al. 2001; Wano
et al. 2001; Buck et al. 1991; Himmelhaus et al. 2000; Bain
et al. 1989; Peterlinz et al. 1996; Jung et al. 2000; Damos
et al. 2005; Bryant et al. 1991). Among all these tech-
niques, electrochemical method is the easiest way to
characterize the electrodesorption process showing a
Fig. 1 a Device configuration
consisting of three layers.
Pictures of b the microfluidic
device and c the device on the
SPR window. Proteins injected
into the channel adsorb on
SAMs and change the incident
light angle of the laser (Dh).
d shows the working principle
of SPR protein sensors (cross
section A–A0 ), and e indicates
one of SPR sensorgram
examples
821
current density (j) versus potential (E) profile. Ellipsometry
is used to evaluate the thickness of SAMs on a metal
electrode (Balasubramanian et al. 2006). SPR is a real time,
in situ optical technique that monitors the incident light
angle needed to excite surface plasmons at the interface of
a thin film of metal and a dielectric layer, such as a SAM.
As the thickness of the dielectric layer increases—for
example, as proteins bind to a SAM—the resonance angle
decreases. Thus, the change in resonance angle shows the
amount of protein adsorbing to a SAM.
2.4.1 Linear sweep voltammetry
To obtain the operating potential of SAM desorption in the
microfluidic device and confirm that the reductive desorp-
tion would not be obscured by HER and the electrode cor-
rosion, we conducted ex situ electrochemical measurement
using linear sweep voltammetry. A custom-made electrodes
set was built; electrodes were placed at the bottom hole of
the cell with a silicone rubber O-ring and the top hole was
tightly fitted with a silicon rubber stopper having a Pt wire
counter and Ag/AgCl reference electrodes. Linear sweep
voltammetry was conducted using a computer-controlled
potentiostat (PGSTAT302N, Eco Chemie).
2.4.2 SPR measurement and surface modification
procedure
All surface modification steps upon electrochemical
desorption on the metal electrode were monitored by SPR
in situ in real time. SPR angle shift was produced by dif-
ferential measurement using two (reference and sensing)
channels in the microfluidic device. As shown in Fig. 1, the
fabricated microfluidic chip was mounted on the SPR
analytical system (Biosensing Instrument Inc.), and the
angle shift was recorded in real time as each solution
inducing surface modification flows through the microflu-
idic channels driven by an external syringe pump. Initially,
123
822
PBS was circulated for 20 min until the angle shift stabi-
lized. Once the angle shift stabilized, the Au electrode in
sample channel was modified with COOH-terminated
SAM of 2 mM 3-mercaptopropionic acid (MPA) solution
in ethanol for 1 h followed by thorough rinsing with PBS.
The first target molecule, streptavidin, was immobilized on
the SAM by activating the carboxylic acid groups of MPA
with freshly prepared 400 mM EDC/100 mM NHS in
water for 15 min. The NHS-esters produced reacted with
amine functions present in the 1-lM streptavidin in 10 mM
NaAc pH 5.3 (sodium acetate) solution. When the protein
immobilization completed, PBS washed the electrode to
remove excess weakly bound proteins. Then, a reductive
potential was applied to desorb the immobilized molecules
on the electrode. The DC potential applied to the Au bot-
tom electrode (cathode) against the top electrode (anode)
for 30 s and the electrode was immediately washed by PBS
to avoid re-adsorption of streptavidin-bound SAM mole-
cules. After complete desorption of the streptavidin-bound
SAM, we formed CH3-terminated SAM at second cycle to
demonstrate reconfigurability. 2 mM 1-butanethiol (BT)
solution in ethanol modified the Au electrode in the sample
channel for 1 h. Following PBS washing, the second target
molecule, fibrinogen, flowed through the channel and
formed hydrophobic bonding with the SAM. The contact
angle measurement showed that the electrode covered by
the SAM had strong hydrophobicity (102°). Finally, the
reductive potential was applied to remove the fibrinogen-
bound SAM.
2.4.3 Evaluation using ellipsometry
Ellipsometry was performed to measure the thickness of
the molecule layers on the electrode before and after the
electrochemical desorption. Also, the thicknesses measured
by ellipsometry were compared with those converted from
SPR signals. Separately prepared Au-coated samples were
used with the ex situ ellipsometry setup.
3 Results and discussion
3.1 Reductive desorption of SAMs on the cathode
There are two challenges to overcome in order to form the
reusable/reconfigurable biosensor in a microfluidic device;
(1) electrochemical reductive desorption of SAMs without
HER and (2) reductive potential to avoid peeling of the
metal electrode. HER is the most challenging problem
during the electrochemical desorption process. Electrolysis
is a process to produce hydrogen at the cathode and oxygen
at the anode when electrodes are in an aqueous solution
and the current runs through the solution. When reductive
123
Microfluid Nanofluid (2009) 7:819–827
potential is applied on the cathode, it is difficult to measure
reductive peak potential, Ep, since electrodesorption of
SAMs and HER occur simultaneously. HER can be fatal in
a microfluidic channel because the bubbles generated by
HER tend to stay on the cathode and interfere with trans-
duction signal. SPR measurements also produce unreliable
data by a tiny bubble in a channel. All baseline data are lost
even after the bubble is removed by the buffer solution.
Electrodesorption and HER were monitored by vol-
tammetry, measuring the current density versus potential
profile in the negative direction. The potential for the
reductive desorption of thiol SAMs depends on the length
of the alkyl chain, the type of terminal groups and the
binding of proteins (Imabayashi et al. 1997; Anderson et al.
2003).
Figure 2 shows linear sweep voltammograms of Au
electrodes coated with SAMs in PBS solution, having
different alkyl chain lengths and different terminal groups.
The DC potential was swept from 0 V to -1.2 V at the rate
of 20 mV/s. Figure 2a1 shows the voltammograms of
COOH(CH2)nSH for n = 2 and 10. Ep of the reductive
desorption of adsorbed thiols appears at -0.7680 V for
n = 2 and -0.8887 V for n = 10 (Fig. 3). On the
other hand, Ep of the corresponding n-alkanethiols,
CH3(CH2)nSH, appears at -0.8047 V for n = 3 and
-0.9072 V for n = 11 as shown in Fig. 2b1 and in Fig. 3.
Two reductive potential peaks are observed in short-chain
SAMs whereas long-chain SAMs show only one peak in
the voltammetry. Among the two peaks, the final reductive
desorption of the SAMs occurs at the second peak poten-
tial. It is shown that Ep of the reductive desorption varies
depending on the length of alkyl chain and the type of
terminal group of SAMs. The longer the alkyl chain is, the
more negative Ep becomes, reflecting the stronger van der
Waals attractive interaction among alkyl chains (Schilardi
et al. 2006). Ep is also dependent on the terminal groups; Ep
of the reductive desorption of COOH(CH2)nSH is more
positive than that of CH3(CH2)nSH. This is due to the
repulsive interaction between the negatively charged
carboxylate groups in a COOH–SAM (Imabayashi et al.
1997).
Figure 2a2, a3 present the voltammograms of covalent
bond of streptavidin to COOH(CH2)nSH by formation of an
amide bond. Reductive desorption peak potentials are more
negatively shifted by 128 mV for n = 2 and by 58 mV for
n = 10 when the surface is modified by streptavidin. Fig-
ure 2b2, b3 show the voltammograms of adsorption of
fibrinogen on CH3(CH2)nSH by hydrophobic attraction.
The desorption peak potentials are also shifted to more
negative values by 92 and 84 mV for n = 2 and 11,
respectively, after the surface is modified by fibrinogen.
Regardless of bonding mechanism of proteins, more
potential is required to desorb the protein-bound SAMs
Microfluid Nanofluid (2009) 7:819–827
Fig. 2 Current density versus potential profiles for the reductive
electrochemical desorption of (a1) x-carboxylic acid alkanethiols
(COOH(CH2)nSH) for n = 2 and 10 (b1) n-alkanethiols
(CH3(CH2)nSH) for n = 3 and 11 (the longer the alkylchain, the
Fig. 3 Reductive peak potential of SH SAMs and protein-bound
SAMs. a COOH(CH2)n. b CH3(CH2)nSH
than that of unmodified SAM. Anderson et al. reported
similar findings, stating that the SAM modified by 1,3- and
1,4-phenylenediamine had more negatively shifted reduc-
tive desorption (Anderson et al. 2003). Figure 2a3, b3
show that protein-bound long-chain SAMs need approxi-
mately -1.0 V to desorb the SAMs and such high potential
seriously damages the opposite top electrode, especially
when the electrode is a thin metal film. Following Sect. 3.2
823
more negative the reductive peak potential) a2, a3 COOH(CH2)nSH
modified by streptavidin for n = 2 and 10, respectively and b2, b3
CH3(CH2)nSH modified by fibrinogen for n = 3 and 11, respectively
addresses this issue. For all different SAMs in Fig. 2, HER
occurs at approximately -1.2 V, which does not overlap
the reductive desorption potential. Thus, short-chain SAMs
have a much less chance to encounter HER.
3.2 Oxidative desorption of SAMs on the anode
Our biosensor experiences both reductive and oxidative
processes on electrodes; reductive process on the cathode
and oxidative process on the anode. Figure 4 shows the
voltammogram of unmodified short-chain SAMs where the
potential is swept between 0 and ?1.2 V. At ?0.9 V,
oxidative desorption of SAMs occurs and Au electrodes
begin to corrode over ?1.0 V. As the potential further
increases, the gold film peels off from the glass substrate.
In order to remove longer chain SAMs bound by proteins
on the cathode (Figs. 2, 3), more than DC 1.0 V is
required, which induces corrosion of gold on the anode.
This is why we use the short-chain SAMs for the reusable/
reconfigurable biosensor to minimize HER on the cathode
and prevent the corrosion of a metal film.
3.3 Real time monitoring of SAM desorption
in a microfluidic device
SPR and ellipsometry were used in concert throughout this
study to verify assembly and removal of molecules in the
microfluidic device. Electrochemical reduction was per-
formed in situ in a two electrode setup and SPR monitored
the molecular interactions in real time. Ellipsometry was
performed ex situ to confirm the assembly of multiple
layers on the bottom Au electrode surface. From SPR angle
shift, proteins’ and SAMs’ thicknesses and their refractive
123
824
Fig. 4 Current density versus potential profiles for the oxidative
desorption of CH3–SAM (square dot) and COOH–SAM (solid line)
indices are calculated according to the method of de Feij-
ter, hence allowing the transformation of a film thickness
parameter into an amount of adsorbed protein per unit area,

reflowed PBS to thoroughly wash unformed SAM residues.
C ng mm2 ¼ d ½ðnf  nm Þ=ðdn=dCÞ; ð2Þ
where d is the film thickness (nm), nf is the refractive index
of the film, nm is the refractive index of the ambient, and
dn/dC is the refractive index increment (De Feijter et al.
1978). This method is based on the assumption that the
refractive index of a protein solution is a linear function of
the protein concentration. This formula for the air/solid
measurement is as follows,

verted values. Reductive desorption of the immobilized
C ng mm2 ¼ Kt; ð3Þ
where K is the density of the protein and SAMs and t is the
thickness (nm). The density of protein are taken to be 1.36
g/cm3 (Balasubramanian et al. 2006; Campbell et al. 2007;
Jung et al. 1998) and the values of SAMs we used are
obtained from the vendor (http://www.sigma-aldrich.com):
0.842 g/cm3 (CH3(CH2)3SH), 0.845 g/cm3 (CH3(CH2)11
SH), 1.218 g/cm3 (COOH(CH2)2SH), 0.792 g/cm3 (COOH
Fig. 5 SPR profiles with schematics of step-by-step procedure of
surface modification.
Injection of COOH–SAM solution. `
COOH–SAM formation. ´ Activation of surface carboxylates. ˆ
Immobilization of streptavidin. ˜ Reductive desorption at -0.9 V. Þ
123
Microfluid Nanofluid (2009) 7:819–827
(CH2)10SH), respectively. For the SPR instrument, an
increase in the plasmon resonance angle of 120 mDeg
corresponds to an average material layer growth of
1 ng/mm2.
Figure 5 shows SPR profiles as a sequence of surface
modification steps. Table 1 summarizes the SPR angles
from Fig. 5, the thicknesses converted from SPR and
ellipsometry results. First, COOH-terminated SAM was
formed by flowing 40 ll of 2 mM MPA in ethanol at the
rate of 10 ll/min. SPR angle shift increased by
1,800 mDeg. When the value stabilized, we stopped the
flowing for 1 h to form SAMs. SAM formation is a two-
step process, where thiol molecules first adsorb on a solid
surface and then rearrange to become a packed monolayer
(Damos et al. 2005). 40–50% of the adsorption occurs very
rapidly within *10 s and over 90% of formation is com-
pleted in 10–100 min (Pan et al. 1996). After 1 h, we
The thickness of the SAM was measured by ellipsometry to
be 4.06 Å, which is in good agreement with SPR; 58.2-
mDeg shift, corresponding to 3.98 Å. Then EDC/NHS
mixture activated carboxylic acid groups of MPA mono-
layer to effectively immobilize streptavidin. Streptavidin
containing amine groups on the surface interacted readily
with the activated MPA intermediates and formed covalent
amide linkages, showing 11.95 Å thick from SPR con-
proteins was performed by applying -0.9 V to the cathode
against Au top electrode for 30 s.
During the desorption process, the constant flow of the
PBS was maintained to avoid re-adsorption of the detached
SAMs. The amplitude of the desorption potential was
based on the data from linear sweep voltammetry mea-
surement. By applying reductive desorption potential
twice, almost complete removal of adlayers was achieved
Injection of CH3–SAM solution. þCH3–SAM formation. ¼Adsorp-
tion of fibrinogen. ½ Reductive desorption at -0.9 V. (a, b) The
enlarged SPR sensorgram of reductive desorption 1 and 2,
respectively.)
Microfluid Nanofluid (2009) 7:819–827 825

Table 1 SPR angle shift and

Adlayer Step no. SPR angle Thickness Thickness
thickness
in Fig. 4 (mDeg) (SPR) (Å) (ellipsometry) (Å)

COOH–SAM ` 58.2 3.98 4.06 ± 0.2

EDC/NHS ´ 88.1 – –
Streptavidin ˆ 283.2 11.95 14.33 ± 0.1
Desorption ˜ 2.8 0.02 0.3 ± 0.08
The thickness measured by CH3–SAM þ 52.8 4.98 5.22 ± 0.2
ellipsometry is in good Fibrinogen ¼ 697.3 42.7 49.5 ± 0.4
agreement with the converted Desorption ½ 1.3 0.01 0.2 ± 0.06
values from the SPR signal

(˜ in Fig. 5; Table 1). We repeated the cycle using CH3- electrodes because the flow rate of 10 ll/min is fast enough
terminated short-chain SAM (CH3(CH2)3SH) and captured to completely remove the dissociated residues out of the
the second target molecule, Fibrinogen, to demonstrate a channel.
reconfigurable biosensor. The bond between fibrinogen and
CH3-terminated SAM was formed by a strong hydrophobic 3.4 Reproducibility
attraction. These phenomena are attributed to the charac-
teristic structure of proteins. Generally, proteins have We performed five measurements of the above procedure in
hydrophobic residues buried within the core of the proteins the same microfluidic device and monitored the change
and their hydrophilic residues facing outside. The adsorp- of the SPR signal (two cycles:COOH–SAM formation
tion of any protein onto a solid surface is a strong function ? streptavidin immobilization ? regeneration ? CH3–
of hydrophobic attraction. Hydrophilic residues form the SAM formation ? fibrinogen adsorption ? regeneration).
orientation of the adsorbed protein and do not take part in Figure 6 and Table 2 show absolute angle values of the SPR
the adsorption process itself (Latour 2005; Choi et al. after each protein modification and regeneration, respec-
2008). Therefore, proteins rapidly adsorb onto a hydro- tively. The binding activity of streptavidin and fibrinogen
phobic surface, unfold and spread their hydrophobic core retained over 95% of the original SPR change signal after 10
over the surface. The solid surface, CH3-terminated SAM, cycles of regeneration. The result indicated that the micro-
employed here showed high hydrophobicity of 102° in fluidic device can be reused with good reproducibility with
contact angle measurement. As a result of fibrinogen an RSD lower than 1.86%.
adsorption, strong bonding occurred between the protein
and the surface. Figure 5 shows that even the PBS flow
does not dissociate the protein-surface binding. The 4 Conclusion
Fibrinogen-bound SAM was also completely desorbed at
-0.9 V (½ in Fig 5 and Table 1). This demonstrates that Many biosensors are designed to be disposable, taking
the sensing surface resets to be reusable/reconfigurable by advantage of low-cost materials and batch fabrication.
undergoing electrochemical desorption of short-chain
SAMs while avoiding HER and electrode corrosion.
One thing to note is that SAM formation and dissocia-
tion may occur on top Au electrode (anode) if high enough
potential is applied. Figure 4 shows that COOH(CH2)3SH
has two oxidative peaks at ?0.6756 and ?0.913 V, while a
desorption peak of CH3(CH2)2SH is shifted to more posi-
tive values, ?0.9502 V. Therefore, DC 0.9 V for the
reductive desorption of SAMs on the cathode does not
completely remove the SAMs formed on the top Au elec-
trode. However, we expect that this imperfect desorption at
the anode has no effect on the reductive desorption or
adsorption of SAM on the cathode since SPR only mea-
sures biomolecular interactions within approximately
200 nm from the cathode. Moreover, 30 s of DC is unlikely Fig. 6 Reproducibility of the SPR response from the protein
to re-adsorb newly detached alkyl molecules on both modifications with regeneration using SAM desorption

123
826 Microfluid Nanofluid (2009) 7:819–827

Table 2 SPR angle shift and thickness of Fig. 6 Barry R, Ivanov D (2004) Microfluidics in biotechnology. J Nano-
biotechnol 2:2
Streptavidin Fibrinogen Bryant MA, Pemberton JE (1991) Surface Raman scattering of self-
assembled monolayers formed from 1-alkanethiols at silver.
SPR angle Thickness SPR angle Thickness
J Am Chem Soc 113:3629–3637
(nm) (nm)
Buck M, Eisert F, Fischer J, Grunze M, Traeger F (1991) Investi-
1 cycle 283 mDeg 1.73 696 mDeg 4.26 gation of self-organizing thiol films by optical second harmonic
generation and X-ray photoelectron spectroscopy. Appl Phys A
2 cycle 287 mDeg 1.76 702 mDeg 4.30 53:552–556
3 cycle 290 mDeg 1.78 705 mDeg 4.32 Campbell CT, Kim G (2007) SPR microscopy and its applications to
4 cycle 293 mDeg 1.79 715 mDeg 4.38 high-throughput analyses of biomolecular binding events and
their kinetics. Biomaterials 28:2380–2392
5 cycle 297 mDeg 1.82 722 mDeg 4.42
Canaria CA, So J, Maloney JR, Yu CJ, Smith JO, Roukes ML, Fraser
Average 290 mDeg 1.78 708 mDeg 4.34 SE, Landsford R (2006) Formation and removal of alkylthiolate
SD 5.39 mDeg 0.034 10.42 mDeg 0.064 self-assembled monolayers on gold in aqueous solutions. Lab
RSD 1.86 1.47 Chip 6:289–295
Choi S, Yong Y, Chae J (2008) Surface plasmon resonance protein
SD standard deviation, RSD relative standard deviation sensor using Vroman effect. Biosens Bioelectron 24:893–899
Damos FS, Luz RCS, Kubota LT (2005) Determination of thickness,
dielectric constant of thiol films, and kinetics of adsorption using
Disposable biosensor chips are much favored for many surface plasmon resonance. Langmuir 21:602–609
applications where it is difficult to wash all residues in de Feijter JA, Benjamins J, Veer FA (1978) Ellipsometry as a tool to
microfluidic channels and sensing surfaces. On the other study the adsorption behavior of synthetic and biopolymers at
the air–water interface. Biopolymers 17:1759–1772
hand, reusable and reconfigurable biosensors are useful Deval J, Umali TA, Lan EH, Bunn B, Ho C (2004) Reconfigurable
where detailed calibration of sensors is required such as in hydrophobic/hydrophilic surfaces in microelectromechanical
biochemical science and analytical chemistry. In the paper, systems (MEMS). J Micromech Microeng 14:91–95
we report a biosensor platform for reusable/reconfigurable Gau J, Lan EH, Dunn B, Ho C, Woo JCS (2001) A MEMS based
amperometric detector for E. Coli bacteria using self-assembled
sensing surfaces using electrochemical reductive desorption monolayers. Biosens Bioelectron 16:745–755
of linker molecules. HER in microfluidic channels and Han Y, Uosaki K (2008) Effects of concentration and temperature on
electrode corrosion are avoided by using short-chain linker the formation process of decanethiol self-assembled monolayer
molecules. The linker molecules bound to proteins are des- on Au(1 1 1) followed by electrochemical reductive desorption.
Electrochim Acta 53:6196–6201
orbed by applying very small DC potential, which are Himmelhaus M, Eisert F, Fischer J, Grunze M (2000) Self-assembly
characterized by ellipsometry, voltammetry, and SPR in real of n-alkanethiol monolayers. A study by IR–visible sum
time. This technique can be comprehensively utilized in all frequency spectroscopy. J Phys Chem B 104:576–584
communities where linker molecules are required to immo- Imabayashi S, Iida M, Hobara D, Feng ZQ, Niki K, Kakiuchi T
(1997) Reductive desorption of carboxylic-acid-terminated
bilize biomolecular receptors. Further research is required to alkanethiol monolayers from Au(111) surfaces. J Electroanal
determine an effective life time of the device while main- Chem 428:33–38
taining reliable formation and desorption of SAMs. Jung LS, Campbell CT (2000) Sticking probabilities in adsorption of
alkanethiols from liquid ethanol solution onto gold. J Phys Chem
B 104:11168–11178
Jung LS, Campbell CT, Chinowsky TM, Mar MN, Yee SS (1998)
Quantitative interpretation of the response of surface plasmon
References resonance sensors to adsorbed films. Langmuir 14:5636–5648
Latour RA Jr (2005) Biomaterials: protein–surface interactions.
Akram M, Stuart MC, Wong DKY (2004) Direct application strategy Encyclopedia of biomaterials and biomedical engineering.
to immobilise a thioctic acid self-assembled monolayer on a gold Taylor and Francis , New York, pp 1–15
electrode. Anal Chim Acta 504:243–251 Lee LYS, Lennox RB (2007) Electrochemical desorption of
Anderson MR, Baltzersen R (2003) Reductive desorption of 11- n-alkylthiol SAMs on polycrystalline gold: studies using a
mercaptoundecanoic acid monolayers modified by covalent ferrocenylalkylthiol probe. Langmuir 23:292–296
attachment of 1,3- and 1,4-phenylenediamine. J Colloid Interf Limbut W, Kanatharana P, Mattiasson B, Asawatreratanakul P,
Sci 263:516–521 Thavarungkul P (2006) A comparative study of capacitive
Asanov AN, Wilson WW, Oldham PB (1998) Regenerable biosensor immunosensors based on self-assembled monolayers formed
platform: a total internal reflection fluorescence cell with from thiourea, thioctic acid, and 3-mercaptopropionic acid.
electrochemical control. Anal Chem 70:1156–1163 Biosens Bioelectron 22:233–240
Bain CD, Troughton EB, Tao YT, Evall J, Whitesides GM, Nuzzo RG Lukaszewski M, Czerwinski A (2006) Dissolution of noble metals
(1989) Formation of monolayer films by the spontaneous and their alloys studied by electrochemical quartz crystal
assembly of organic thiols from solution onto gold. J Am Chem microbalance. J Electroanal Chem 589:38–45
Soc 111:321–335 Minunni M, Tombelli S, Gullotto A, Luzi E, Mascini M (2004)
Balasubramanian S, Revzin A, Simnian A (2006) Electrochemical Development of biosensors with aptamers as bio-recognition
desorption of proteins from gold electrode surface. Electroanal- element: the case of HIV-1 Tat protein. Biosens Bioelectron
ysis 18:1885–1892 20:1149–1156

123
Microfluid Nanofluid (2009) 7:819–827
Mirsky VM (2002) New electroanalytical applications of self-
assembled monolayers. Trends Anal Chem 21:439–450
Oyamatsu D, Fujita T, Arimoto S, Munakata H, Matsumoto H,
Kuwabata S (2008) Electrochemical desorption of a self-
assembled monolayer of alkanethiol in ionic liquids. J Electro-
anal Chem 615:110–116
Pan W, Durning CJ, Turro NJ (1996) Kinetics of alkanethiol
adsorption on gold. Langmuir 12:4469–4473
Peterlinz KA, Georgiadis R (1996) In situ kinetics of self-assembly by
surface plasmon resonance spectroscopy. Langmuir 12:4731–
4740
Pyun JC, Kim SD, Chung JW (2005) New immobilization method for
immunoaffinity biosensors by using thiolated proteins. Anal
Biochem 347:227–233
Qu D, Morin M (2001) An EQCM study of the oxidative deposition
of alkylthiolates on gold. J Electroanal Chem 517:45–53
Santini JT, Richards AC, Scheidt R, Chima MJ, Langer R (2000)
Microchips as controlled drug-delivery devices. Angew Chem
Int Ed 39:2396–2407
Schilardi PL, Dip P, do Santos Claro PC, Benitez GA, Fonticelli MH,
Azzaroni O, Salvarezza RC (2006) Electrochemical deposition
onto self-assembled monolayers: new insights into micro- and
nanofabrication. Chem Eur J 12:38–49
Vericat C, Andreasen G, Vela ME, Martin H, Salvarezza RC (2001)
Following transformation in self-assembled alkanethiol monolay-
ers on Au(111) by in situ scanning tunneling microscopy. J Chem
Phys 115:6672–6678
827
Wang Y, Luo J, Chen H, He Q, Gan N, Li T (2008) A microchip-
based flow injection-amperometry system with mercaptopropi-
onic acid modified electroless gold microelectrode for the
selective determination of dopamine. Anal Chim Acta 625:
180–187
Wano W, Uosaki K (2001) In Situ, real-time monitoring of the
reductive desorption process of self-assembled monolayers of
hexanethiol on Au(111) surfaces in acidic and alkaline aqueous
solutions by scanning tunneling microscopy. Langmuir 17:8224–
8228
Widrig CA, Chung C, Porter MD (1991) The electrochemical
desorption of n-alkanethiol monolayers from polycrystalline
Au and Ag electrodes. J Electroanal Chem 310:335–359
Yang DF, Al-Maznai H, Morin M (1997) Vibrational study of the fast
reductive and the slow oxidative desorptions of a nonanethiol
self-assembled monolayer from a Au(111) single crystal elec-
trode. J Phys Chem B 101:1158–1166
Zhang X, Wu Y, Tu Y, Liu S (2008) A reusable electrochemical
immunosensor for carcinoembryonic antigen via molecular
recognition of glycoprotein antibody by phenylboronic acid
self-assembly layer on gold. Analyst 133:485–492
123